2.2. Performance Guide

2.2.1. Linux 09.01.00 Performance Guide

Read This First

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

Name

Description

AM68 SK

AM68 SK Evaluation Module with ARM running at 2GHz, DDR data rate 2666 MT/S, L3 Cache size 3MB

AM69 SK

AM69 SK Evaluation Module with ARM running at 2GHz, DDR data rate 2666 MT/S, L3 Cache size 3MB

J721e EVM

J721e Evaluation Module rev E2 with ARM running at 2GHz, DDR data rate 4266 MT/S, L3 Cache size 3MB

J7200 EVM

J7200 Evaluation Module rev E1 with ARM running at 2GHz, DDR data rate 2666 MT/S, L3 Cache size 3MB

J721S2 EVM

J721S2 Evaluation Module rev E2 with ARM running at 2GHz, DDR data rate 2666 MT/S, L3 Cache size 3MB

J784S4 EVM

J784S4 Evaluation Module Beta rev E1 with ARM running at 2GHz, DDR data rate 2666 MT/S, L3 Cache size 1MB

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 http://e2e.ti.com/ or http://support.ti.com/

2.2.1.1. System Benchmarks

2.2.1.1.1. 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 http://lmbench.sourceforge.net/whatis_lmbench.html and http://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

Benchmarks

am68_sk-fs: perf

am69_sk-fs: perf

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

af_unix_sock_stream_latency (microsec)

15.56

18.82

17.83

15.71

15.95

15.85

af_unix_socket_stream_bandwidth (MBs)

2072.79

3520.26

1626.69

2043.75

1896.16

3458.34

bw_file_rd-io-1mb (MB/s)

3013.75

3577.82

2254.79

2922.91

2818.49

3684.21

bw_file_rd-o2c-1mb (MB/s)

1260.81

323.89

1168.42

1557.63

1270.88

982.16

bw_mem-bcopy-16mb (MB/s)

3395.23

3180.91

2380.60

2858.67

3456.84

3226.78

bw_mem-bcopy-1mb (MB/s)

4492.70

10149.37

3305.79

5123.34

4535.15

9825.33

bw_mem-bcopy-2mb (MB/s)

3589.38

4738.47

2496.88

4054.74

3639.01

4982.21

bw_mem-bcopy-4mb (MB/s)

3414.72

3865.48

2350.18

3734.83

3477.05

3954.13

bw_mem-bcopy-8mb (MB/s)

3396.55

3337.04

2357.45

3104.38

3439.87

3414.91

bw_mem-bzero-16mb (MB/s)

10491.80

10856.66

2334.06

9657.46

10533.25

10851.14

bw_mem-bzero-1mb (MB/s)

8602.66 (min 4492.70, max 12712.62)

11957.50 (min 10149.37, max 13765.62)

4109.68 (min 3305.79, max 4913.56)

8869.34 (min 5123.34, max 12615.33)

8611.39 (min 4535.15, max 12687.63)

11792.99 (min 9825.33, max 13760.64)

bw_mem-bzero-2mb (MB/s)

7443.35 (min 3589.38, max 11297.31)

9201.31 (min 4738.47, max 13664.15)

2706.70 (min 2496.88, max 2916.51)

8111.44 (min 4054.74, max 12168.14)

7420.70 (min 3639.01, max 11202.39)

9347.84 (min 4982.21, max 13713.46)

bw_mem-bzero-4mb (MB/s)

7013.34 (min 3414.72, max 10611.96)

8083.25 (min 3865.48, max 12301.01)

2411.57 (min 2350.18, max 2472.95)

7806.54 (min 3734.83, max 11878.25)

7055.79 (min 3477.05, max 10634.53)

8115.36 (min 3954.13, max 12276.58)

bw_mem-bzero-8mb (MB/s)

6967.38 (min 3396.55, max 10538.20)

7339.27 (min 3337.04, max 11341.49)

2355.89 (min 2354.33, max 2357.45)

7370.37 (min 3104.38, max 11636.36)

7010.07 (min 3439.87, max 10580.26)

7352.19 (min 3414.91, max 11289.47)

bw_mem-cp-16mb (MB/s)

2112.21

2426.08

989.18

1592.36

2157.21

2403.85

bw_mem-cp-1mb (MB/s)

7732.12 (min 2526.23, max 12938.01)

8698.15 (min 3663.00, max 13733.29)

3058.84 (min 1257.18, max 4860.49)

7217.42 (min 1790.33, max 12644.51)

7491.85 (min 2401.63, max 12582.06)

8711.93 (min 3688.09, max 13735.77)

bw_mem-cp-2mb (MB/s)

6702.26 (min 2067.54, max 11336.98)

8191.09 (min 2708.19, max 13673.98)

1952.08 (min 1003.51, max 2900.65)

6851.59 (min 1444.30, max 12258.88)

6744.02 (min 2197.11, max 11290.93)

8247.34 (min 2778.74, max 13715.94)

bw_mem-cp-4mb (MB/s)

6321.63 (min 2029.43, max 10613.83)

7599.41 (min 2841.92, max 12356.90)

1717.41 (min 978.95, max 2455.87)

6762.76 (min 1624.92, max 11900.60)

6279.20 (min 1938.92, max 10619.47)

7794.84 (min 2853.07, max 12736.60)

bw_mem-cp-8mb (MB/s)

6316.27 (min 2087.14, max 10545.39)

6952.10 (min 2475.63, max 11428.57)

1637.61 (min 987.53, max 2287.68)

6574.27 (min 1541.72, max 11606.82)

6357.83 (min 2116.12, max 10599.54)

6904.54 (min 2449.48, max 11359.60)

bw_mem-fcp-16mb (MB/s)

3362.05

3140.95

2399.16

2830.86

3381.59

3170.83

bw_mem-fcp-1mb (MB/s)

8344.74 (min 3976.86, max 12712.62)

10022.08 (min 6278.54, max 13765.62)

4095.83 (min 3278.09, max 4913.56)

8198.88 (min 3782.42, max 12615.33)

8272.29 (min 3856.94, max 12687.63)

10249.47 (min 6738.30, max 13760.64)

bw_mem-fcp-2mb (MB/s)

7391.16 (min 3485.00, max 11297.31)

8715.67 (min 3767.19, max 13664.15)

2716.12 (min 2515.72, max 2916.51)

8062.13 (min 3956.12, max 12168.14)

7366.94 (min 3531.49, max 11202.39)

8995.77 (min 4278.07, max 13713.46)

bw_mem-fcp-4mb (MB/s)

7007.82 (min 3403.68, max 10611.96)

7981.67 (min 3662.33, max 12301.01)

2432.89 (min 2392.82, max 2472.95)

7884.65 (min 3891.05, max 11878.25)

7001.90 (min 3369.27, max 10634.53)

8022.60 (min 3768.61, max 12276.58)

bw_mem-fcp-8mb (MB/s)

6945.78 (min 3353.36, max 10538.20)

7315.48 (min 3289.47, max 11341.49)

2383.26 (min 2354.33, max 2412.18)

7364.18 (min 3091.99, max 11636.36)

6996.13 (min 3412.00, max 10580.26)

7287.67 (min 3285.87, max 11289.47)

bw_mem-frd-16mb (MB/s)

4156.38

3706.28

6312.47

4801.20

4147.76

3757.63

bw_mem-frd-1mb (MB/s)

4716.20 (min 3976.86, max 5455.54)

7073.40 (min 6278.54, max 7868.25)

5002.49 (min 3278.09, max 6726.89)

4480.45 (min 3782.42, max 5178.47)

4487.02 (min 3856.94, max 5117.10)

7342.42 (min 6738.30, max 7946.54)

bw_mem-frd-2mb (MB/s)

4107.50 (min 3485.00, max 4730.00)

4107.72 (min 3767.19, max 4448.25)

4461.85 (min 2515.72, max 6407.97)

4882.40 (min 3956.12, max 5808.68)

4132.14 (min 3531.49, max 4732.79)

4696.12 (min 4278.07, max 5114.16)

bw_mem-frd-4mb (MB/s)

3796.45 (min 3403.68, max 4189.21)

3913.78 (min 3662.33, max 4165.22)

4342.92 (min 2392.82, max 6293.02)

4762.92 (min 3891.05, max 5634.79)

3778.88 (min 3369.27, max 4188.48)

4074.89 (min 3768.61, max 4381.16)

bw_mem-frd-8mb (MB/s)

3760.02 (min 3353.36, max 4166.67)

3670.05 (min 3289.47, max 4050.63)

4354.21 (min 2412.18, max 6296.24)

4243.69 (min 3091.99, max 5395.38)

3781.41 (min 3412.00, max 4150.81)

3760.46 (min 3285.87, max 4235.04)

bw_mem-fwr-16mb (MB/s)

10529.78

10827.27

2319.18

9637.10

10548.87

10788.94

bw_mem-fwr-1mb (MB/s)

9196.78 (min 5455.54, max 12938.01)

10800.77 (min 7868.25, max 13733.29)

5793.69 (min 4860.49, max 6726.89)

8911.49 (min 5178.47, max 12644.51)

8849.58 (min 5117.10, max 12582.06)

10841.16 (min 7946.54, max 13735.77)

bw_mem-fwr-2mb (MB/s)

8033.49 (min 4730.00, max 11336.98)

9061.12 (min 4448.25, max 13673.98)

4654.31 (min 2900.65, max 6407.97)

9033.78 (min 5808.68, max 12258.88)

8011.86 (min 4732.79, max 11290.93)

9415.05 (min 5114.16, max 13715.94)

bw_mem-fwr-4mb (MB/s)

7401.52 (min 4189.21, max 10613.83)

8261.06 (min 4165.22, max 12356.90)

4374.45 (min 2455.87, max 6293.02)

8767.70 (min 5634.79, max 11900.60)

7403.98 (min 4188.48, max 10619.47)

8558.88 (min 4381.16, max 12736.60)

bw_mem-fwr-8mb (MB/s)

7356.03 (min 4166.67, max 10545.39)

7739.60 (min 4050.63, max 11428.57)

4291.96 (min 2287.68, max 6296.24)

8501.10 (min 5395.38, max 11606.82)

7375.18 (min 4150.81, max 10599.54)

7797.32 (min 4235.04, max 11359.60)

bw_mem-rd-16mb (MB/s)

4879.54

4424.17

6558.27

5117.54

4889.98

4507.68

bw_mem-rd-1mb (MB/s)

7372.07 (min 5745.06, max 8999.08)

15998.07 (min 14764.13, max 17232.00)

13177.14 (min 11446.01, max 14908.26)

7757.05 (min 6240.25, max 9273.84)

7848.88 (min 6117.38, max 9580.38)

16288.46 (min 15269.95, max 17306.96)

bw_mem-rd-2mb (MB/s)

4418.29 (min 2980.63, max 5855.94)

5099.74 (min 3926.19, max 6273.28)

4025.35 (min 973.55, max 7077.14)

3813.60 (min 1367.99, max 6259.20)

4379.04 (min 3065.13, max 5692.94)

5741.09 (min 4901.96, max 6580.22)

bw_mem-rd-4mb (MB/s)

3625.22 (min 2268.22, max 4982.21)

4438.31 (min 3909.43, max 4967.18)

3712.52 (min 787.87, max 6637.17)

3791.61 (min 1314.71, max 6268.50)

3670.04 (min 2354.33, max 4985.75)

4716.87 (min 4098.36, max 5335.37)

bw_mem-rd-8mb (MB/s)

3541.03 (min 2236.51, max 4845.55)

4372.35 (min 3792.67, max 4952.03)

3662.75 (min 748.71, max 6576.78)

3583.48 (min 1315.79, max 5851.16)

3571.44 (min 2252.89, max 4889.98)

4478.75 (min 3750.00, max 5207.49)

bw_mem-rdwr-16mb (MB/s)

2141.90

2639.83

752.09

1919.62

2167.73

2657.37

bw_mem-rdwr-1mb (MB/s)

3511.32 (min 2526.23, max 4496.40)

6608.86 (min 3663.00, max 9554.71)

2534.81 (min 1257.18, max 3812.43)

3156.33 (min 1790.33, max 4522.33)

3945.05 (min 2401.63, max 5488.47)

6594.42 (min 3688.09, max 9500.75)

bw_mem-rdwr-2mb (MB/s)

2357.27 (min 2067.54, max 2647.00)

3516.05 (min 2708.19, max 4323.90)

995.58 (min 987.65, max 1003.51)

1945.89 (min 1444.30, max 2447.48)

2611.99 (min 2197.11, max 3026.86)

3557.39 (min 2778.74, max 4336.04)

bw_mem-rdwr-4mb (MB/s)

2135.64 (min 2029.43, max 2241.84)

3092.09 (min 2841.92, max 3342.25)

885.48 (min 792.00, max 978.95)

2016.74 (min 1624.92, max 2408.55)

2141.97 (min 1938.92, max 2345.01)

3155.15 (min 2853.07, max 3457.22)

bw_mem-rdwr-8mb (MB/s)

2140.67 (min 2087.14, max 2194.19)

2820.72 (min 2475.63, max 3165.81)

871.48 (min 755.43, max 987.53)

1936.36 (min 1541.72, max 2331.00)

2176.94 (min 2116.12, max 2237.76)

2842.43 (min 2449.48, max 3235.37)

bw_mem-wr-16mb (MB/s)

2200.22

3061.03

742.15

1820.87

2225.93

3030.88

bw_mem-wr-1mb (MB/s)

5120.73 (min 4496.40, max 5745.06)

13393.36 (min 9554.71, max 17232.00)

7629.22 (min 3812.43, max 11446.01)

5381.29 (min 4522.33, max 6240.25)

5802.93 (min 5488.47, max 6117.38)

13403.86 (min 9500.75, max 17306.96)

bw_mem-wr-2mb (MB/s)

2813.82 (min 2647.00, max 2980.63)

4125.05 (min 3926.19, max 4323.90)

980.60 (min 973.55, max 987.65)

1907.74 (min 1367.99, max 2447.48)

3046.00 (min 3026.86, max 3065.13)

4619.00 (min 4336.04, max 4901.96)

bw_mem-wr-4mb (MB/s)

2255.03 (min 2241.84, max 2268.22)

3625.84 (min 3342.25, max 3909.43)

789.94 (min 787.87, max 792.00)

1861.63 (min 1314.71, max 2408.55)

2349.67 (min 2345.01, max 2354.33)

3777.79 (min 3457.22, max 4098.36)

bw_mem-wr-8mb (MB/s)

2215.35 (min 2194.19, max 2236.51)

3479.24 (min 3165.81, max 3792.67)

752.07 (min 748.71, max 755.43)

1823.40 (min 1315.79, max 2331.00)

2245.33 (min 2237.76, max 2252.89)

3492.69 (min 3235.37, max 3750.00)

bw_mmap_rd-mo-1mb (MB/s)

8828.72

12916.14

8124.08

8448.12

6874.77

12917.93

bw_mmap_rd-o2c-1mb (MB/s)

1159.42

311.53

1587.02

1529.57

1171.87

930.23

bw_pipe (MB/s)

830.52

670.16

671.55

795.56

801.99

907.75

bw_unix (MB/s)

2072.79

3520.26

1626.69

2043.75

1896.16

3458.34

lat_connect (us)

35.55

34.20

35.22

34.38

35.57

34.83

lat_ctx-2-128k (us)

3.17

5.39

3.30

3.25

3.29

3.12

lat_ctx-2-256k (us)

2.49

4.86

2.56

2.55

2.56

2.42

lat_ctx-4-128k (us)

3.28

4.01

3.36

3.22

3.30

3.82

lat_ctx-4-256k (us)

2.04

2.35

2.15

2.14

2.18

3.53

lat_fs-0k (num_files)

530.00

562.00

435.00

557.00

544.00

450.00

lat_fs-10k (num_files)

160.00

218.00

150.00

201.00

201.00

239.00

lat_fs-1k (num_files)

269.00

265.00

277.00

270.00

287.00

296.00

lat_fs-4k (num_files)

301.00

309.00

283.00

297.00

251.00

309.00

lat_mem_rd-stride128-sz1000k (ns)

12.01

5.65

10.48

11.15

13.40

5.71

lat_mem_rd-stride128-sz125k (ns)

5.57

5.65

5.57

5.57

5.57

5.65

lat_mem_rd-stride128-sz250k (ns)

5.57

5.65

5.58

5.57

5.57

5.65

lat_mem_rd-stride128-sz31k (ns)

3.79

4.75

2.00

4.67

3.34

4.77

lat_mem_rd-stride128-sz50 (ns)

2.00

2.00

2.00

2.00

2.00

2.00

lat_mem_rd-stride128-sz500k (ns)

5.57

5.65

5.57

5.57

5.57

5.65

lat_mem_rd-stride128-sz62k (ns)

5.57

5.20

5.58

5.12

5.12

5.20

lat_mmap-1m (us)

34.00

27.00

27.00

33.00

27.00

27.00

lat_ops-double-add (ns)

1.96

1.96

1.96

1.96

1.96

1.96

lat_ops-double-div (ns)

9.01

9.01

9.01

9.01

9.01

9.01

lat_ops-double-mul (ns)

2.00

2.00

2.00

2.00

2.00

2.00

lat_ops-float-add (ns)

1.96

1.96

1.96

1.96

1.96

1.96

lat_ops-float-div (ns)

5.50

5.51

5.51

5.50

5.50

5.50

lat_ops-float-mul (ns)

2.00

2.00

2.02

2.00

2.01

2.00

lat_ops-int-add (ns)

0.50

0.50

0.50

0.50

0.50

0.50

lat_ops-int-bit (ns)

0.33

0.33

0.33

0.33

0.33

0.33

lat_ops-int-div (ns)

4.00

4.00

4.01

4.00

4.00

4.00

lat_ops-int-mod (ns)

4.67

4.67

4.67

4.67

4.67

4.67

lat_ops-int-mul (ns)

1.52

1.52

1.52

1.52

1.52

1.52

lat_ops-int64-add (ns)

0.50

0.50

0.50

0.50

0.50

0.50

lat_ops-int64-bit (ns)

0.33

0.33

0.33

0.33

0.33

0.33

lat_ops-int64-div (ns)

3.00

3.00

3.00

3.00

3.00

3.00

lat_ops-int64-mod (ns)

5.68

5.67

5.67

5.67

5.67

5.67

lat_ops-int64-mul (ns)

2.52

2.52

2.52

2.52

2.52

2.52

lat_pagefault (us)

0.52

0.51

0.51

0.50

0.53

0.51

lat_pipe (us)

11.90

12.08

11.98

11.86

11.56

12.38

lat_proc-exec (us)

435.91

405.00

441.50

379.20

442.54

425.50

lat_proc-fork (us)

355.31

364.00

376.43

323.88

356.21

349.27

lat_proc-proccall (us)

0.00

0.00

0.00

0.00

0.00

0.00

lat_select (us)

10.89

13.09

10.87

10.81

10.80

10.90

lat_sem (us)

1.59

1.17

1.43

1.29

1.49

1.21

lat_sig-catch (us)

3.06

3.24

3.14

3.05

3.06

3.09

lat_sig-install (us)

0.50

0.50

0.50

0.50

0.50

0.50

lat_sig-prot (us)

0.48

0.30

0.45

0.52

0.49

0.45

lat_syscall-fstat (us)

1.13

1.21

1.14

1.12

1.15

1.17

lat_syscall-null (us)

0.37

0.37

0.37

0.37

0.37

0.37

lat_syscall-open (us)

110.37

2866.00

105.18

164.79

151.26

153.57

lat_syscall-read (us)

0.47

0.46

0.47

0.47

0.47

0.46

lat_syscall-stat (us)

1.43

1.57

1.47

1.40

1.41

1.45

lat_syscall-write (us)

0.42

0.43

0.42

0.44

0.42

0.42

lat_tcp (us)

0.78

0.79

0.78

0.78

0.78

0.79

lat_unix (us)

15.56

18.82

17.83

15.71

15.95

15.85

latency_for_0.50_mb_block_size (nanosec)

5.57

5.65

5.57

5.57

5.57

5.65

latency_for_1.00_mb_block_size (nanosec)

6.01 (min 0.00, max 12.01)

2.82 (min 0.00, max 5.65)

5.24 (min 0.00, max 10.48)

5.58 (min 0.00, max 11.15)

6.70 (min 0.00, max 13.40)

2.86 (min 0.00, max 5.71)

pipe_bandwidth (MBs)

830.52

670.16

671.55

795.56

801.99

907.75

pipe_latency (microsec)

11.90

12.08

11.98

11.86

11.56

12.38

procedure_call (microsec)

0.00

0.00

0.00

0.00

0.00

0.00

select_on_200_tcp_fds (microsec)

10.89

13.09

10.87

10.81

10.80

10.90

semaphore_latency (microsec)

1.59

1.17

1.43

1.29

1.49

1.21

signal_handler_latency (microsec)

0.50

0.50

0.50

0.50

0.50

0.50

signal_handler_overhead (microsec)

3.06

3.24

3.14

3.05

3.06

3.09

tcp_ip_connection_cost_to_localhost (microsec)

35.55

34.20

35.22

34.38

35.57

34.83

tcp_latency_using_localhost (microsec)

0.78

0.79

0.78

0.78

0.78

0.79

Table: LM Bench Metrics

2.2.1.1.2. Dhrystone

Dhrystone is a core only benchmark that runs from warm L1 caches in all modern processors. It scales linearly with clock speed. For standard ARM cores the DMIPS/MHz score will be identical with the same compiler and flags.

Execute the benchmark with the following:

runDhrystone

Benchmarks

am68_sk-fs: perf

am69_sk-fs: perf

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

cpu_clock (MHz)

2000.00

2000.00

2000.00

2000.00

2000.00

2000.00

dhrystone_per_mhz (DMIPS/MHz)

5.20

4.40

5.20

4.40

4.10

4.40

dhrystone_per_second (DhrystoneP)

18181818.00

15384615.00

18181818.00

15384615.00

14285714.00

15384615.00

Table: Dhrystone Benchmark

2.2.1.1.3. Whetstone

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

Execute the benchmark with the following:

runWhetstone

Benchmarks

am68_sk-fs: perf

am69_sk-fs: perf

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

whetstone (MIPS)

10000.00

10000.00

10000.00

10000.00

10000.00

10000.00

Table: Whetstone Benchmark

2.2.1.1.4. Linpack

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

Benchmarks

am68_sk-fs: perf

am69_sk-fs: perf

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

linpack (Kflops)

2592175.00

2486122.00

2391734.00

2459408.00

2372725.00

2601996.00

Table: Linpack Benchmark

2.2.1.1.5. 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

Benchmarks

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

assignment (Iterations)

30.86

30.81

30.80

30.73

fourier (Iterations)

53004.00

64902.00

65145.00

50203.00

fp_emulation (Iterations)

214.60

214.63

214.64

214.62

huffman (Iterations)

2381.20

2376.80

2383.80

2378.00

idea (Iterations)

7996.80

7997.00

7996.80

7996.70

lu_decomposition (Iterations)

1375.00

1364.40

1360.50

1360.80

neural_net (Iterations)

29.49

29.78

28.96

29.91

numeric_sort (Iterations)

855.31

848.33

854.56

853.11

string_sort (Iterations)

413.85

425.25

417.58

417.49

Table: NBench Benchmarks

2.2.1.1.6. 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

Benchmarks

am68_sk-fs: perf

am69_sk-fs: perf

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

add (MB/s)

6364.50

6216.20

5459.10

5380.80

6340.50

6150.90

copy (MB/s)

6930.30

7145.70

4779.80

5631.00

7001.30

7047.50

scale (MB/s)

7063.10

7159.50

4836.90

5504.10

7133.30

7073.70

triad (MB/s)

6372.60

6213.70

5460.50

5361.60

6350.40

6150.40

Table: Stream

2.2.1.1.7. 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.

Benchmarks

am68_sk-fs: perf

am69_sk-fs: perf

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

cjpeg-rose7-preset (workloads/)

82.64

82.64

82.64

83.33

81.97

82.64

core (workloads/)

0.77

0.78

0.78

0.78

0.78

0.78

coremark-pro ()

2504.00

2518.83

2424.25

2465.22

2475.03

2499.36

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

80.52

79.87

79.62

80.52

80.91

81.57

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

2.47

2.47

2.45

2.46

2.49

2.47

nnet_test (workloads/)

3.64

3.84

3.58

3.61

3.65

3.59

parser-125k (workloads/)

11.11

10.87

11.11

10.99

11.11

10.87

radix2-big-64k (workloads/)

283.29

274.35

216.64

250.56

253.94

267.95

sha-test (workloads/)

158.73

158.73

158.73

156.25

156.25

158.73

zip-test (workloads/)

47.62

50.00

47.62

47.62

47.62

50.00

Table: CoreMarkPro

2.2.1.1.8. 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.

Benchmarks

am68_sk-fs: perf

am69_sk-fs: perf

j7200-evm: perf

j721e-idk-gw: perf

j721s2-evm: perf

j784s4-evm: perf

4m-check (workloads/)

901.55

1035.20

935.28

931.79

888.73

1050.42

4m-check-reassembly (workloads/)

153.37

202.84

120.92

147.28

148.81

200.40

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

99.21

115.74

93.28

101.22

94.34

116.28

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

41.52

36.25

40.82

44.05

41.61

38.61

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

2.67

4.67

2.70

2.72

2.67

4.89

4m-cmykw2 (workloads/)

314.96

604.23

313.48

317.46

313.48

598.80

4m-cmykw2-rotatew2 (workloads/)

58.54

48.00

58.88

62.24

59.23

53.86

4m-reassembly (workloads/)

125.95

152.44

108.70

129.87

126.58

150.15

4m-rotatew2 (workloads/)

71.53

52.85

70.87

74.02

70.87

56.53

4m-tcp-mixed (workloads/)

250.00

253.97

280.70

280.70

266.67

271.19

4m-x264w2 (workloads/)

2.75

4.78

2.76

2.76

2.72

5.10

idct-4m (workloads/)

35.00

35.15

34.94

35.09

34.98

35.11

idct-4mw1 (workloads/)

35.01

35.11

34.94

35.08

34.98

35.14

ippktcheck-4m (workloads/)

893.82

1035.63

926.27

928.68

851.79

1042.10

ippktcheck-4mw1 (workloads/)

913.74

1025.43

951.29

916.09

884.02

1028.38

ipres-4m (workloads/)

168.73

208.04

158.06

179.86

163.40

202.98

ipres-4mw1 (workloads/)

169.68

202.98

155.44

181.60

165.20

201.61

md5-4m (workloads/)

45.50

47.57

45.87

48.61

44.68

47.87

md5-4mw1 (workloads/)

46.27

47.44

45.68

48.85

46.17

47.30

rgbcmyk-4m (workloads/)

163.13

164.20

163.13

163.93

163.13

164.07

rgbcmyk-4mw1 (workloads/)

163.13

163.80

163.13

163.80

161.81

163.93

rotate-4ms1 (workloads/)

52.25

55.07

52.36

54.64

51.76

54.70

rotate-4ms1w1 (workloads/)

52.14

55.07

52.47

54.95

51.49

55.01

rotate-4ms64 (workloads/)

52.85

55.56

52.91

55.37

52.19

55.49

rotate-4ms64w1 (workloads/)

52.80

55.62

53.08

55.25

52.74

55.37

x264-4mq (workloads/)

1.42

1.41

1.42

1.43

1.42

1.43

x264-4mqw1 (workloads/)

1.41

1.43

1.42

1.44

1.41

1.43

Table: Multibench

2.2.1.2. Boot-time Measurement

2.2.1.2.1. Boot media: MMCSD

Boot Configuration

am68_sk-fs: boot time (sec)

am69_sk-fs: boot time (sec)

j7200-evm: boot time (sec)

j721e-idk-gw: boot time (sec)

j721s2-evm: boot time (sec)

j784s4-evm: boot time (sec)

Kernel boot time test when bootloader, kernel and sdk-rootfs are in mmc-sd

13.81 (min 13.69, max 13.89)

14.55 (min 14.47, max 14.68)

13.55 (min 13.28, max 14.12)

17.38 (min 17.05, max 18.16)

18.58 (min 18.49, max 18.79)

16.22 (min 15.57, max 18.18)

Kernel boot time test when init is /bin/sh and bootloader, kernel and sdk-rootfs are in mmc-sd

4.58 (min 4.57, max 4.61)

5.22 (min 5.20, max 5.24)

4.31 (min 4.28, max 4.33)

5.03 (min 5.01, max 5.07)

5.33 (min 5.32, max 5.33)

6.41 (min 6.29, max 6.45)

Table: Boot time MMC/SD

2.2.1.3. ALSA SoC Audio Driver

  1. Access type - RW_INTERLEAVED

  2. Channels - 2

  3. Format - S16_LE

  4. Period size - 64

Sampling Rate (Hz)

j721e-idk-gw: Throughput (bits/sec)

j721e-idk-gw: CPU Load (%)

11025

352793.00

0.16

16000

511992.00

0.28

22050

705576.00

0.23

24000

705583.00

0.27

32000

1023980.00

0.36

44100

1411175.00

0.45

48000

1535974.00

0.27

88200

2822351.00

0.83

96000

3071946.00

1.12

Table: Audio Capture


Sampling Rate (Hz)

j721e-idk-gw: Throughput (bits/sec)

j721e-idk-gw: CPU Load (%)

11025

352935.00

0.14

16000

512198.00

0.19

22050

705872.00

0.23

24000

705872.00

0.24

32000

1024396.00

0.32

44100

1411745.00

0.42

48000

1536594.00

0.45

88200

2823489.00

0.79

96000

2964662.00

1.25

Table: Audio Playback


2.2.1.4. Graphics SGX/RGX Driver

2.2.1.4.1. Glmark2

Run Glmark2 and capture performance reported (Score). All display outputs (HDMI, Displayport and/or LCD) are connected when running these tests

Benchmark

am68_sk-fs: Score

am69_sk-fs: Score

j721e-idk-gw: Score

j721s2-evm: Score

j784s4-evm: Score

Glmark2-DRM

56.00

56.00

28.00

28.00

Glmark2-Wayland

1332.00

1401.00

1180.00

1286.00

1333.00

Table: Glmark2


2.2.1.5. 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

2.2.1.5.1. CPSW/CPSW2g/CPSW3g Ethernet Driver

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

  • CPSW3g: AM64x

TCP Bidirectional Throughput

Command Used

j721e-idk-gw: THROUGHPUT (Mbits/sec)

j721e-idk-gw: CPU Load % (LOCAL_CPU_UTIL)

j721s2-evm: THROUGHPUT (Mbits/sec)

j721s2-evm: CPU Load % (LOCAL_CPU_UTIL)

j784s4-evm: THROUGHPUT (Mbits/sec)

j784s4-evm: 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

1834.33

73.50

1864.35

75.70

1855.53

18.66

Table: CPSW TCP Bidirectional Throughput


TCP Bidirectional Throughput Interrupt Pacing

Command Used

j721e-idk-gw: THROUGHPUT (Mbits/sec)

j721e-idk-gw: CPU Load % (LOCAL_CPU_UTIL)

j721s2-evm: THROUGHPUT (Mbits/sec)

j721s2-evm: 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

1859.60

34.09

1873.43

38.54

Table: CPSW TCP Bidirectional Throughput Interrupt Pacing


UDP Throughput

Frame Size(bytes)

j721e-idk-gw: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

j721e-idk-gw: THROUGHPUT (Mbits/sec)

j721e-idk-gw: Packets Per Second (kPPS)

j721e-idk-gw: CPU Load % (LOCAL_CPU_UTIL)

j721s2-evm: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

j721s2-evm: THROUGHPUT (Mbits/sec)

j721s2-evm: Packets Per Second (kPPS)

j721s2-evm: CPU Load % (LOCAL_CPU_UTIL)

64

18.00

32.39

225.00

88.25

18.00

3.63

25.00

18.05

128

82.00

146.16

223.00

88.80

82.00

138.80

212.00

90.72

256

210.00

368.14

219.00

87.76

210.00

27.55

16.00

12.52

1024

978.00

675.97

86.00

35.83

978.00

936.05

120.00

55.72

1518

1472.00

592.23

50.00

23.48

1472.00

955.98

81.00

42.40

Table: CPSW UDP Egress Throughput |

Frame Size(bytes)

j721e-idk-gw: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

j721e-idk-gw: THROUGHPUT (Mbits/sec)

j721e-idk-gw: Packets Per Second (kPPS)

j721e-idk-gw: CPU Load % (LOCAL_CPU_UTIL)

j721s2-evm: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

j721s2-evm: THROUGHPUT (Mbits/sec)

j721s2-evm: Packets Per Second (kPPS)

j721s2-evm: CPU Load % (LOCAL_CPU_UTIL)

64

18.00

7.26

50.00

18.15

18.00

1.28

9.00

6.56

128

82.00

8.20

13.00

5.59

82.00

20.53

31.00

21.86

256

210.00

18.31

11.00

4.62

210.00

76.44

46.00

29.70

1024

978.00

107.19

14.00

8.80

978.00

180.73

23.00

15.57

1518

1472.00

216.68

18.00

13.52

1472.00

956.98

81.00

64.71

Table: CPSW UDP Ingress Throughput (0% loss)


Frame Size(bytes)

j721e-idk-gw: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

j721e-idk-gw: THROUGHPUT (Mbits/sec)

j721e-idk-gw: Packets Per Second (kPPS)

j721e-idk-gw: CPU Load % (LOCAL_CPU_UTIL)

j721e-idk-gw: Packet Loss %

j721s2-evm: UDP Datagram Size(bytes) (LOCAL_SEND_SIZE)

j721s2-evm: THROUGHPUT (Mbits/sec)

j721s2-evm: Packets Per Second (kPPS)

j721s2-evm: CPU Load % (LOCAL_CPU_UTIL)

j721s2-evm: Packet Loss %

64

18.00

59.08

410.00

88.04

1.78

18.00

18.61

129.00

71.03

0.04

128

82.00

261.28

398.00

90.62

1.68

82.00

86.81

132.00

72.16

4.34

256

210.00

669.82

399.00

92.76

1.82

210.00

234.27

139.00

80.60

0.06

1024

978.00

928.84

119.00

83.35

0.03

978.00

930.94

119.00

84.56

0.61

1518

1472.00

949.65

81.00

60.31

0.02

1472.00

956.98

81.00

64.71

0.00

Table: CPSW UDP Ingress Throughput (possible loss)


2.2.1.6. PCIe Driver

2.2.1.6.1. PCIe-ETH

TCP Window Size(Kbytes)

j7200-evm: Bandwidth (Mbits/sec)

j721e-idk-gw: Bandwidth (Mbits/sec)

8

225.60

268.00

16

225.60

243.20

32

357.60

367.20

64

561.60

595.20

128

737.60

751.20

256

808.00

808.00

Table: PCI Ethernet

2.2.1.6.2. PCIe-NVMe-SSD

2.2.1.6.2.1. J721E-IDK-GW

Buffer size (bytes)

j721e-idk-gw: Write EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Write EXT4 CPU Load (%)

j721e-idk-gw: Read EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Read EXT4 CPU Load (%)

1m

720.00

13.11

1521.00

6.09

4m

721.00

11.86

1523.00

5.01

4k

187.00

48.64

166.00

36.34

256k

740.00

12.56

1520.00

11.87

  • Filesize used is: 10G

  • FIO command options: –ioengine=libaio –iodepth=4 –numjobs=1 –direct=1 –runtime=60 –time_based

  • Platform: Speed 8GT/s, Width x2

  • SSD being used: PLEXTOR PX-128M8PeY

2.2.1.6.2.2. J7200-EVM

Buffer size (bytes)

j7200-evm: Write EXT4 Throughput (Mbytes/sec)

j7200-evm: Write EXT4 CPU Load (%)

j7200-evm: Read EXT4 Throughput (Mbytes/sec)

j7200-evm: Read EXT4 CPU Load (%)

1m

722.00

14.86

1524.00

6.79

4m

718.00

13.94

1524.00

5.74

4k

184.00

48.57

166.00

36.55

256k

742.00

14.18

1521.00

11.99

  • Filesize used is: 10G

  • FIO command options: –ioengine=libaio –iodepth=4 –numjobs=1 –direct=1 –runtime=60 –time_based

  • Platform: Speed 8GT/s, Width x2

  • SSD being used: PLEXTOR PX-128M8PeY

2.2.1.6.2.3. J721S2-EVM

Buffer size (bytes)

j721s2-evm: Write EXT4 Throughput (Mbytes/sec)

j721s2-evm: Write EXT4 CPU Load (%)

j721s2-evm: Read EXT4 Throughput (Mbytes/sec)

j721s2-evm: Read EXT4 CPU Load (%)

1m

744.00

17.41

781.00

5.54

4m

744.00

15.43

782.00

5.31

4k

184.00

51.52

284.00

51.48

256k

744.00

16.66

780.00

7.99

  • Filesize used is: 10G

  • FIO command options: –ioengine=libaio –iodepth=4 –numjobs=1 –direct=1 –runtime=60 –time_based

  • Platform: Speed 8GT/s, Width x2

  • SSD being used: PLEXTOR PX-128M8PeY

2.2.1.7. OSPI Flash Driver

2.2.1.7.1. J721E-IDK-GW

2.2.1.7.1.1. RAW

File size (Mbytes)

j721e-idk-gw: Raw Read Throughput (Mbytes/sec)

50

38.17

2.2.1.7.2. J7200-EVM

2.2.1.7.2.1. RAW

File size (Mbytes)

j7200-evm: Raw Read Throughput (Mbytes/sec)

50

208.33

2.2.1.7.3. J784S4-EVM

2.2.1.7.3.1. RAW

File size (Mbytes)

j784s4-evm: Raw Read Throughput (Mbytes/sec)

50

263.16

2.2.1.8. UBoot QSPI/OSPI Driver

2.2.1.8.1. J721E-IDK-GW

File size (bytes in hex)

j721e-idk-gw: Write Throughput (Kbytes/sec)

j721e-idk-gw: Read Throughput (Kbytes/sec)

400000

1539.85

37577.98

800000

1541.59

39009.52

1000000

1542.89

39766.99

2000000

1542.75

40206.13

2.2.1.8.2. J7200-EVM

File size (bytes in hex)

j7200-evm: Write Throughput (Kbytes/sec)

j7200-evm: Read Throughput (Kbytes/sec)

400000

345.83

204800.00

800000

349.46

240941.18

1000000

348.70

277694.92

2000000

343.68

300623.85

2.2.1.8.3. J721S2-EVM

File size (bytes in hex)

j721s2-evm: Write Throughput (Kbytes/sec)

j721s2-evm: Read Throughput (Kbytes/sec)

400000

377.86

204800.00

800000

379.70

248242.42

1000000

381.30

277694.92

2000000

375.68

300623.85

2.2.1.9. UBoot UFS Driver


2.2.1.9.1. J721E-IDK-GW


File size (bytes in hex)

j721e-idk-gw: Write Throughput (Kbytes/sec)

j721e-idk-gw: Read Throughput (Kbytes/sec)

400000

93090.91

341333.33

800000

97523.81

481882.35

1000000

91530.73

606814.81

2.2.1.10. 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.


2.2.1.10.1. J7200-EVM


Buffer size (bytes)

j7200-evm: Write EXT4 Throughput (Mbytes/sec)

j7200-evm: Write EXT4 CPU Load (%)

j7200-evm: Read EXT4 Throughput (Mbytes/sec)

j7200-evm: Read EXT4 CPU Load (%)

1m

56.20

1.52

314.00

1.92

4m

56.40

1.27

314.00

1.31

4k

54.50

25.02

55.90

23.44

256k

56.30

1.67

314.00

3.42

2.2.1.10.2. J721E-IDK-GW


Buffer size (bytes)

j721e-idk-gw: Write EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Write EXT4 CPU Load (%)

j721e-idk-gw: Read EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Read EXT4 CPU Load (%)

1m

59.50

1.25

175.00

1.02

4m

59.10

1.10

175.00

0.66

4k

49.80

23.01

56.60

23.65

256k

58.90

1.70

174.00

1.93

2.2.1.10.3. J721S2-EVM


Buffer size (bytes)

j721s2-evm: Write EXT4 Throughput (Mbytes/sec)

j721s2-evm: Write EXT4 CPU Load (%)

j721s2-evm: Read EXT4 Throughput (Mbytes/sec)

j721s2-evm: Read EXT4 CPU Load (%)

1m

45.20

3.56

301.00

3.76

4m

45.30

3.35

278.00

3.54

4k

5.27

5.12

36.20

19.00

256k

36.40

3.18

284.00

5.53

2.2.1.10.4. J784S4-EVM


Buffer size (bytes)

j784s4-evm: Write EXT4 Throughput (Mbytes/sec)

j784s4-evm: Write EXT4 CPU Load (%)

j784s4-evm: Read EXT4 Throughput (Mbytes/sec)

j784s4-evm: Read EXT4 CPU Load (%)

1m

90.80

0.45

173.00

0.30

4m

97.50

0.50

150.00

0.20

4k

63.50

8.85

93.20

10.70

256k

90.80

0.59

172.00

0.55

2.2.1.11. UBoot EMMC Driver


2.2.1.11.1. J7200-EVM


File size (bytes in hex)

j7200-evm: Write Throughput (Kbytes/sec)

j7200-evm: Read Throughput (Kbytes/sec)

2000000

56109.59

309132.08

4000000

56496.55

321254.90

2.2.1.11.2. J721E-IDK-GW


File size (bytes in hex)

j721e-idk-gw: Write Throughput (Kbytes/sec)

j721e-idk-gw: Read Throughput (Kbytes/sec)

2000000

59686.70

173375.66

4000000

60513.39

176646.90

2.2.1.11.3. J721S2-EVM


File size (bytes in hex)

j721s2-evm: Write Throughput (Kbytes/sec)

j721s2-evm: Read Throughput (Kbytes/sec)

2000000

60014.65

303407.41

4000000

60346.22

303407.41

2.2.1.11.4. J784S4-EVM


File size (bytes in hex)

j784s4-evm: Write Throughput (Kbytes/sec)

j784s4-evm: Read Throughput (Kbytes/sec)

2000000

98402.40

159067.96

4000000

97090.37

186181.82

2.2.1.12. MMC/SD 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.


2.2.1.12.1. J7200-EVM


Buffer size (bytes)

j7200-evm: Write EXT4 Throughput (Mbytes/sec)

j7200-evm: Write EXT4 CPU Load (%)

j7200-evm: Read EXT4 Throughput (Mbytes/sec)

j7200-evm: Read EXT4 CPU Load (%)

1m

19.40

0.77

86.30

0.68

4m

19.90

0.66

86.40

0.53

4k

5.42

3.35

16.70

7.25

256k

19.40

0.88

84.30

1.19

2.2.1.12.2. J721S2-EVM


Buffer size (bytes)

j721s2-evm: Write EXT4 Throughput (Mbytes/sec)

j721s2-evm: Write EXT4 CPU Load (%)

j721s2-evm: Read EXT4 Throughput (Mbytes/sec)

j721s2-evm: Read EXT4 CPU Load (%)

1m

20.40

2.83

86.20

2.42

4m

20.70

2.77

86.40

2.82

4k

5.64

5.67

16.60

9.83

256k

20.00

2.48

83.80

3.17

2.2.1.12.3. J784S4-EVM


Buffer size (bytes)

j784s4-evm: Write EXT4 Throughput (Mbytes/sec)

j784s4-evm: Write EXT4 CPU Load (%)

j784s4-evm: Read EXT4 Throughput (Mbytes/sec)

j784s4-evm: Read EXT4 CPU Load (%)

1m

18.50

0.15

86.90

0.21

4m

19.00

0.18

86.90

0.18

4k

4.65

0.83

16.50

2.18

256k

18.10

0.17

84.80

0.38

2.2.1.12.4. J721E-IDK-GW


Buffer size (bytes)

j721e-idk-gw: Write EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Write EXT4 CPU Load (%)

j721e-idk-gw: Read EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Read EXT4 CPU Load (%)

1m

18.80

0.70

43.50

0.48

4m

19.40

0.59

43.50

0.45

4k

4.75

2.76

13.70

5.93

256k

18.50

0.70

42.80

0.71


The performance numbers were captured using the following:

  • SanDisk 8GB MicroSDHC Class 10 Memory Card

  • Partition was mounted with async option


2.2.1.13. UBoot MMC/SD Driver


2.2.1.13.1. J721E-IDK-GW

File size (bytes in hex)

j721e-idk-gw: Write Throughput (Kbytes/sec)

j721e-idk-gw: Read Throughput (Kbytes/sec)

400000

27125.83

35617.39

800000

28743.86

40554.46

1000000

36167.77

43574.47

2.2.1.13.2. J7200-EVM

File size (bytes in hex)

j7200-evm: Write Throughput (Kbytes/sec)

j7200-evm: Read Throughput (Kbytes/sec)

400000

19230.05

71859.65

800000

21903.74

81108.91

1000000

14615.52

85780.10

The performance numbers were captured using the following:

  • SanDisk 8GB MicroSDHC Class 10 Memory Card

2.2.1.13.3. J721S2-EVM

File size (bytes in hex)

j721s2-evm: Write Throughput (Kbytes/sec)

j721s2-evm: Read Throughput (Kbytes/sec)

400000

34133.33

60235.29

800000

21671.96

73142.86

1000000

20924.65

82331.66

The performance numbers were captured using the following:

  • SanDisk 8GB MicroSDHC Class 10 Memory Card

2.2.1.13.4. J784S4-EVM

File size (bytes in hex)

j784s4-evm: Write Throughput (Kbytes/sec)

j784s4-evm: Read Throughput (Kbytes/sec)

400000

24380.95

56888.89

800000

17886.46

70620.69

1000000

18347.14

80313.73

The performance numbers were captured using the following:

  • SanDisk 8GB MicroSDHC Class 10 Memory Card


2.2.1.14. USB Driver

2.2.1.14.1. USB Host Controller

Warning

IMPORTANT: For Mass-storage applications, 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.


Setup : Inateck ASM1153E USB hard disk is connected to usb0 port. File read/write performance data on usb0 port is captured.


2.2.1.14.1.1. J721E-IDK-GW

Buffer size (bytes)

j721e-idk-gw: Write EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Write EXT4 CPU Load (%)

j721e-idk-gw: Read EXT4 Throughput (Mbytes/sec)

j721e-idk-gw: Read EXT4 CPU Load (%)

1m

402.00

8.33

426.00

2.42

4m

405.00

7.11

425.00

2.34

4k

30.90

24.50

61.70

28.20

256k

351.00

12.18

404.00

4.72


2.2.1.14.2. USB Device Controller

Number of Blocks

j721e-idk-gw: Throughput (MB/sec)

j721s2-evm: Throughput (MB/sec)

150

42.40

32.80

Table: USBDEVICE HIGHSPEED SLAVE READ THROUGHPUT



Number of Blocks

j721e-idk-gw: Throughput (MB/sec)

j721s2-evm: Throughput (MB/sec)

150

42.60

29.90

Table: USBDEVICE HIGHSPEED SLAVE WRITE THROUGHPUT



2.2.1.15. CRYPTO Driver

2.2.1.15.1. OpenSSL Performance

Algorithm

Buffer Size (in bytes)

am68_sk-fs: throughput (KBytes/Sec)

am69_sk-fs: throughput (KBytes/Sec)

j721e-idk-gw: throughput (KBytes/Sec)

j721s2-evm: throughput (KBytes/Sec)

j784s4-evm: throughput (KBytes/Sec)

aes-128-cbc

1024

46429.53

45751.30

49285.46

46068.05

45405.18

aes-128-cbc

16

899.54

896.65

965.25

873.54

688.66

aes-128-cbc

16384

186493.61

172474.37

194794.84

183850.33

183298.73

aes-128-cbc

256

13901.14

13697.19

14749.70

13769.05

13555.80

aes-128-cbc

64

3633.73

3591.83

3860.18

3516.78

3462.95

aes-128-cbc

8192

153777.49

147262.12

161376.94

152171.86

150792.87

aes-128-ecb

1024

46818.65

44624.21

49023.32

46254.42

45046.44

aes-128-ecb

16

909.09

681.99

951.10

897.44

691.49

aes-128-ecb

16384

188459.69

187711.49

196504.23

187411.11

172862.12

aes-128-ecb

256

14010.37

10779.90

14685.01

13238.87

13818.28

aes-128-ecb

64

3613.55

2715.75

3786.73

3571.09

2784.11

aes-128-ecb

8192

154271.74

154959.87

165019.65

155901.95

135113.39

aes-192-cbc

1024

45590.87

45221.21

48857.77

45721.60

43852.12

aes-192-cbc

16

908.77

893.10

961.37

900.54

843.21

aes-192-cbc

16384

175876.78

169558.02

182856.36

176062.46

174729.90

aes-192-cbc

256

13746.01

13827.50

14706.94

13150.55

13602.13

aes-192-cbc

64

3624.64

3574.81

3836.20

3602.20

3517.42

aes-192-cbc

8192

146046.98

145697.45

153318.74

148149.59

144788.14

aes-192-ecb

1024

46128.81

43304.62

48797.70

45928.45

45101.74

aes-192-ecb

16

909.42

705.83

952.48

869.00

669.34

aes-192-ecb

16384

182621.53

181600.26

186772.14

184232.62

180240.38

aes-192-ecb

256

13871.62

12962.39

14603.09

13866.41

11489.54

aes-192-ecb

64

3624.51

3164.12

3799.83

3612.59

2712.49

aes-192-ecb

8192

150522.54

150000.98

156499.97

149656.92

148531.88

aes-256-cbc

1024

45124.95

36672.17

48109.23

45196.97

43773.27

aes-256-cbc

16

906.81

665.95

958.40

868.80

819.19

aes-256-cbc

16384

164719.27

163119.10

168564.05

164620.97

163856.38

aes-256-cbc

256

13724.76

10673.07

14615.81

13721.26

13660.50

aes-256-cbc

64

3626.05

2970.82

3816.23

3614.87

3540.99

aes-256-cbc

8192

139332.27

138343.77

144236.54

137336.15

138859.86

aes-256-ecb

1024

45514.75

45401.43

48356.35

45915.48

44286.98

aes-256-ecb

16

907.07

668.95

950.36

907.41

779.80

aes-256-ecb

16384

167832.23

170366.29

173850.62

169590.78

169749.16

aes-256-ecb

256

13864.96

13832.87

14673.07

13498.88

13754.88

aes-256-ecb

64

3630.02

3580.22

3791.19

3584.51

3547.41

aes-256-ecb

8192

142980.44

142453.42

148744.87

143480.15

142060.20

des3

1024

39280.64

39346.18

41344.68

38780.93

38940.67

des3

16

906.81

856.34

964.57

886.84

823.60

des3

16384

95901.01

96485.38

97086.12

95917.40

96266.92

des3

256

13052.33

13075.29

13924.35

13061.21

12959.74

des3

64

3618.90

3584.28

3846.66

3593.13

3550.76

des3

8192

87111.00

87569.75

88896.85

87192.92

87545.17

sha1

1024

58287.79

56244.22

59582.81

59465.39

58433.54

sha1

16

971.70

941.21

994.48

1005.63

979.65

sha1

16384

466239.49

460155.56

460565.16

461651.97

466337.79

sha1

256

15351.38

14636.97

15500.03

15607.98

15352.15

sha1

64

3890.01

3716.97

3922.92

3873.47

3886.74

sha1

8192

310583.30

308401.49

316347.73

319629.99

318354.77

sha256

1024

58359.13

56272.21

57865.22

57578.50

57274.71

sha256

16

982.96

945.89

967.27

990.46

960.04

sha256

16384

465906.35

448664.92

459806.04

456873.30

455010.99

sha256

256

15389.18

14776.75

15104.26

15438.08

15057.83

sha256

64

3863.55

3746.39

3846.72

3906.03

3792.98

sha256

8192

314160.47

302336.68

311252.31

315692.37

310941.01

sha512

1024

46239.06

45265.58

47265.11

46406.31

45240.32

sha512

16

987.65

953.66

1006.12

991.33

948.78

sha512

16384

149990.06

149946.37

151273.47

149908.14

150017.37

sha512

256

14336.60

14068.39

14751.57

14560.77

13920.60

sha512

64

3937.83

3831.42

4029.18

3834.90

3780.39

sha512

8192

130927.27

130034.35

131989.50

130588.67

129982.46


Algorithm

am68_sk-fs: CPU Load

am69_sk-fs: CPU Load

j721e-idk-gw: CPU Load

j721s2-evm: CPU Load

j784s4-evm: CPU Load

aes-128-cbc

34.00

34.00

34.00

34.00

33.00

aes-128-ecb

34.00

33.00

35.00

34.00

34.00

aes-192-cbc

33.00

33.00

34.00

33.00

33.00

aes-192-ecb

34.00

33.00

35.00

34.00

33.00

aes-256-cbc

33.00

32.00

34.00

33.00

33.00

aes-256-ecb

33.00

33.00

34.00

33.00

33.00

des3

30.00

29.00

30.00

30.00

30.00

sha1

98.00

98.00

99.00

99.00

99.00

sha256

98.00

98.00

99.00

99.00

99.00

sha512

98.00

98.00

99.00

99.00

99.00
Listed for each algorithm are the code snippets used to run each

benchmark test.

::

time -v openssl speed -elapsed -evp aes-128-cbc

2.2.1.15.2. IPSec Software Performance

Algorithm

j721e-idk-gw: Throughput (Mbps)

j721e-idk-gw: Packets/Sec

j721e-idk-gw: CPU Load

3des

215.30

19.00

33.54

aes128

650.30

58.00

57.70

aes192

639.60

57.00

57.70

aes256

626.50

55.00

57.58

2.2.1.16. DCAN Driver

Performance and Benchmarks not available in this release.

2.2.2. RT-linux 09.01.00 Performance Guide

Read This First

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

Name

Description

J7200 EVM

J7200 Evaluation Module rev E2 with ARM running at 2GHz, DDR data rate 2666 MT/S, L3 Cache size 3MB

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 http://e2e.ti.com/ or http://support.ti.com/

2.2.2.1. System Benchmarks

2.2.2.1.1. 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 http://lmbench.sourceforge.net/whatis_lmbench.html and http://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

Benchmarks

j7200-evm: perf

af_unix_sock_stream_latency (microsec)

20.81

af_unix_socket_stream_bandwidth (MBs)

1623.79

bw_file_rd-io-1mb (MB/s)

2711.01

bw_file_rd-o2c-1mb (MB/s)

1301.36

bw_mem-bcopy-16mb (MB/s)

2353.63

bw_mem-bcopy-1mb (MB/s)

3405.22

bw_mem-bcopy-2mb (MB/s)

2473.06

bw_mem-bcopy-4mb (MB/s)

2380.48

bw_mem-bcopy-8mb (MB/s)

2365.11

bw_mem-bzero-16mb (MB/s)

2341.58

bw_mem-bzero-1mb (MB/s)

4145.11 (min 3405.22, max 4885.00)

bw_mem-bzero-2mb (MB/s)

2709.56 (min 2473.06, max 2946.05)

bw_mem-bzero-4mb (MB/s)

2421.39 (min 2380.48, max 2462.30)

bw_mem-bzero-8mb (MB/s)

2357.13 (min 2349.14, max 2365.11)

bw_mem-cp-16mb (MB/s)

988.08

bw_mem-cp-1mb (MB/s)

3038.13 (min 1286.06, max 4790.19)

bw_mem-cp-2mb (MB/s)

1968.99 (min 1015.06, max 2922.91)

bw_mem-cp-4mb (MB/s)

1723.52 (min 992.68, max 2454.36)

bw_mem-cp-8mb (MB/s)

1662.28 (min 989.12, max 2335.43)

bw_mem-fcp-16mb (MB/s)

2396.64

bw_mem-fcp-1mb (MB/s)

4130.11 (min 3375.21, max 4885.00)

bw_mem-fcp-2mb (MB/s)

2746.68 (min 2547.31, max 2946.05)

bw_mem-fcp-4mb (MB/s)

2442.05 (min 2421.80, max 2462.30)

bw_mem-fcp-8mb (MB/s)

2380.85 (min 2349.14, max 2412.55)

bw_mem-frd-16mb (MB/s)

6225.68

bw_mem-frd-1mb (MB/s)

5010.20 (min 3375.21, max 6645.19)

bw_mem-frd-2mb (MB/s)

4486.22 (min 2547.31, max 6425.13)

bw_mem-frd-4mb (MB/s)

4271.93 (min 2421.80, max 6122.06)

bw_mem-frd-8mb (MB/s)

4304.17 (min 2412.55, max 6195.79)

bw_mem-fwr-16mb (MB/s)

2320.86

bw_mem-fwr-1mb (MB/s)

5717.69 (min 4790.19, max 6645.19)

bw_mem-fwr-2mb (MB/s)

4674.02 (min 2922.91, max 6425.13)

bw_mem-fwr-4mb (MB/s)

4288.21 (min 2454.36, max 6122.06)

bw_mem-fwr-8mb (MB/s)

4265.61 (min 2335.43, max 6195.79)

bw_mem-rd-16mb (MB/s)

6508.47

bw_mem-rd-1mb (MB/s)

9989.53 (min 7710.24, max 12268.82)

bw_mem-rd-2mb (MB/s)

4022.61 (min 975.45, max 7069.77)

bw_mem-rd-4mb (MB/s)

3706.22 (min 791.14, max 6621.30)

bw_mem-rd-8mb (MB/s)

3609.94 (min 748.43, max 6471.44)

bw_mem-rdwr-16mb (MB/s)

752.09

bw_mem-rdwr-1mb (MB/s)

4606.50 (min 1286.06, max 7926.93)

bw_mem-rdwr-2mb (MB/s)

1001.20 (min 987.33, max 1015.06)

bw_mem-rdwr-4mb (MB/s)

894.79 (min 796.89, max 992.68)

bw_mem-rdwr-8mb (MB/s)

873.82 (min 758.51, max 989.12)

bw_mem-wr-16mb (MB/s)

745.64

bw_mem-wr-1mb (MB/s)

7818.59 (min 7710.24, max 7926.93)

bw_mem-wr-2mb (MB/s)

981.39 (min 975.45, max 987.33)

bw_mem-wr-4mb (MB/s)

794.02 (min 791.14, max 796.89)

bw_mem-wr-8mb (MB/s)

753.47 (min 748.43, max 758.51)

bw_mmap_rd-mo-1mb (MB/s)

9830.38

bw_mmap_rd-o2c-1mb (MB/s)

1413.93

bw_pipe (MB/s)

644.50

bw_unix (MB/s)

1623.79

lat_connect (us)

41.38

lat_ctx-2-128k (us)

3.25

lat_ctx-2-256k (us)

2.57

lat_ctx-4-128k (us)

3.23

lat_ctx-4-256k (us)

2.25

lat_fs-0k (num_files)

439.00

lat_fs-10k (num_files)

129.00

lat_fs-1k (num_files)

228.00

lat_fs-4k (num_files)

238.00

lat_mem_rd-stride128-sz1000k (ns)

11.24

lat_mem_rd-stride128-sz125k (ns)

5.57

lat_mem_rd-stride128-sz250k (ns)

5.57

lat_mem_rd-stride128-sz31k (ns)

4.66

lat_mem_rd-stride128-sz50 (ns)

2.00

lat_mem_rd-stride128-sz500k (ns)

5.58

lat_mem_rd-stride128-sz62k (ns)

5.57

lat_mmap-1m (us)

38.00

lat_ops-double-add (ns)

1.96

lat_ops-double-div (ns)

9.01

lat_ops-double-mul (ns)

2.00

lat_ops-float-add (ns)

1.96

lat_ops-float-div (ns)

5.51

lat_ops-float-mul (ns)

2.01

lat_ops-int-add (ns)

0.50

lat_ops-int-bit (ns)

0.33

lat_ops-int-div (ns)

4.01

lat_ops-int-mod (ns)

4.68

lat_ops-int-mul (ns)

1.52

lat_ops-int64-add (ns)

0.50

lat_ops-int64-bit (ns)

0.33

lat_ops-int64-div (ns)

3.01

lat_ops-int64-mod (ns)

5.68

lat_ops-int64-mul (ns)

2.52

lat_pagefault (us)

0.55

lat_pipe (us)

11.11

lat_proc-exec (us)

493.67

lat_proc-fork (us)

417.69

lat_proc-proccall (us)

0.00

lat_select (us)

12.48

lat_sem (us)

1.76

lat_sig-catch (us)

2.83

lat_sig-install (us)

0.49

lat_sig-prot (us)

0.44

lat_syscall-fstat (us)

1.60

lat_syscall-null (us)

0.38

lat_syscall-open (us)

150.97

lat_syscall-read (us)

0.49

lat_syscall-stat (us)

2.10

lat_syscall-write (us)

0.42

lat_tcp (us)

0.78

lat_unix (us)

20.81

latency_for_0.50_mb_block_size (nanosec)

5.58

latency_for_1.00_mb_block_size (nanosec)

5.62 (min 0.00, max 11.24)

pipe_bandwidth (MBs)

644.50

pipe_latency (microsec)

11.11

procedure_call (microsec)

0.00

select_on_200_tcp_fds (microsec)

12.48

semaphore_latency (microsec)

1.76

signal_handler_latency (microsec)

0.49

signal_handler_overhead (microsec)

2.83

tcp_ip_connection_cost_to_localhost (microsec)

41.38

tcp_latency_using_localhost (microsec)

0.78

Table: LM Bench Metrics

2.2.2.1.2. Dhrystone

Dhrystone is a core only benchmark that runs from warm L1 caches in all modern processors. It scales linearly with clock speed. For standard ARM cores the DMIPS/MHz score will be identical with the same compiler and flags.

Benchmarks

j7200-evm: perf

cpu_clock (MHz)

2000.00

dhrystone_per_mhz (DMIPS/MHz)

5.70

dhrystone_per_second (DhrystoneP)

20000000.00

Table: Dhrystone Benchmark

2.2.2.1.3. Whetstone

Benchmarks

j7200-evm: perf

whetstone (MIPS)

10000.00

Table: Whetstone Benchmark

2.2.2.1.4. Linpack

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

Benchmarks

j7200-evm: perf

linpack (Kflops)

2618867.00

Table: Linpack Benchmark

2.2.2.1.5. 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

Benchmarks

j7200-evm: perf

assignment (Iterations)

30.76

fourier (Iterations)

50468.00

fp_emulation (Iterations)

214.60

huffman (Iterations)

2382.70

idea (Iterations)

7996.00

lu_decomposition (Iterations)

1338.50

neural_net (Iterations)

30.07

numeric_sort (Iterations)

853.41

string_sort (Iterations)

416.14

Table: NBench Benchmarks

2.2.2.1.6. 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.

Benchmarks

j7200-evm: perf

add (MB/s)

5421.70

copy (MB/s)

4767.00

scale (MB/s)

4815.60

triad (MB/s)

5426.20

Table: Stream

2.2.2.1.7. 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.

Benchmarks

j7200-evm: perf

cjpeg-rose7-preset (workloads/)

81.97

core (workloads/)

0.78

coremark-pro ()

2435.28

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

80.65

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

2.46

nnet_test (workloads/)

3.65

parser-125k (workloads/)

10.99

radix2-big-64k (workloads/)

225.33

sha-test (workloads/)

156.25

zip-test (workloads/)

47.62

Table: CoreMarkPro

2.2.2.1.8. 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.

Benchmarks

j7200-evm: perf

4m-check (workloads/)

904.16

4m-check-reassembly (workloads/)

118.77

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

90.58

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

40.71

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

2.68

4m-cmykw2 (workloads/)

309.12

4m-cmykw2-rotatew2 (workloads/)

59.35

4m-reassembly (workloads/)

104.17

4m-rotatew2 (workloads/)

70.27

4m-tcp-mixed (workloads/)

280.70

4m-x264w2 (workloads/)

2.73

idct-4m (workloads/)

34.87

idct-4mw1 (workloads/)

34.83

ippktcheck-4m (workloads/)

892.86

ippktcheck-4mw1 (workloads/)

899.93

ipres-4m (workloads/)

157.23

ipres-4mw1 (workloads/)

157.23

md5-4m (workloads/)

44.17

md5-4mw1 (workloads/)

44.29

rgbcmyk-4m (workloads/)

162.34

rgbcmyk-4mw1 (workloads/)

162.34

rotate-4ms1 (workloads/)

51.71

rotate-4ms1w1 (workloads/)

51.60

rotate-4ms64 (workloads/)

52.30

rotate-4ms64w1 (workloads/)

52.41

x264-4mq (workloads/)

1.41

x264-4mqw1 (workloads/)

1.42

Table: Multibench

2.2.2.2. Boot-time Measurement

2.2.2.2.1. Boot media: MMCSD

Boot Configuration

j7200-evm: boot time (sec)

Kernel boot time test when bootloader, kernel and sdk-rootfs are in mmc-sd

15.32 (min 15.19, max 15.43)

Kernel boot time test when init is /bin/sh and bootloader, kernel and sdk-rootfs are in mmc-sd

4.12 (min 4.08, max 4.18)

Table: Boot time MMC/SD

2.2.2.3. PCIe Driver

2.2.2.3.1. PCIe-ETH

TCP Window Size(Kbytes)

j7200-evm: Bandwidth (Mbits/sec)

8

226.40

16

224.80

32

381.60

64

564.80

128

733.60

256

800.00

Table: PCI Ethernet

2.2.2.4. Cyclic Test

# cyclictest –mlockall –smp –priority=80 –interval=200 –distance=0 policy: fifo: loadavg: 0.24 0.27 0.22 1/169 679

T: 0 ( 678) P:80 I:200 C: 13973 Min: 3 Act: 3 Avg: 3 Max: 24 T: 1 ( 679) P:80 I:200 C: 13923 Min: 3 Act: 3 Avg: 3 Max: 13

2.2.2.5. DCAN Driver

Performance and Benchmarks not available in this release.