Welcome to Embench-LiteX tester’s results!

This page presents the Embench-tester results. The data was obtained by simulating a LiteX based SoC design instantiating the tested core. All the cores where tested in LiteX’ standard configuration.

Table 1 Relative tests results: [%]

cv32e40p_standard_32_False

cv32e41p_standard_32_False

cva5_standard_32_False

femtorv_standard_32_False

firev_standard_32_False

ibex_standard_32_False

lm32_standard_32_False

marocchino_standard_32_False

minerva_standard_32_False

mor1kx_standard_32_False

naxriscv_standard_32_False

neorv32_standard_32_False

picorv32_standard_32_False

vexriscv_standard_32_False

vexriscv_smp_standard_32_False

blackparrot_sim_32_False

microwatt_standard+ghdl_32_False

serv_standard_32_False

aha-mont64

46.0

41.0

82.0

4.0

7.0

46.0

45.0

52.0

7.0

55.0

117.0

31.0

14.0

74.0

71.0

131.0

155.0

0.0

crc32

41.0

36.0

92.0

11.0

25.0

38.0

58.0

55.0

9.0

70.0

40.0

29.0

15.0

74.0

72.0

92.0

68.0

1.0

cubic

24.0

22.0

46.0

2.0

2.0

24.0

19.0

24.0

6.0

29.0

15.0

10.0

8.0

26.0

23.0

238.0

99.0

0.0

edn

42.0

37.0

90.0

1.0

2.0

38.0

62.0

55.0

44.0

60.0

24.0

31.0

10.0

87.0

77.0

90.0

52.0

0.0

huffbench

46.0

44.0

115.0

31.0

44.0

47.0

82.0

61.0

21.0

88.0

35.0

39.0

28.0

112.0

96.0

119.0

76.0

3.0

matmult-int

39.0

34.0

90.0

3.0

3.0

37.0

52.0

39.0

60.0

56.0

21.0

33.0

13.0

82.0

62.0

91.0

44.0

0.0

md5sum

65.0

56.0

153.0

31.0

56.0

67.0

84.0

76.0

19.0

107.0

64.0

50.0

32.0

125.0

116.0

137.0

115.0

3.0

minver

38.0

32.0

84.0

1.0

2.0

37.0

27.0

29.0

8.0

34.0

27.0

15.0

13.0

60.0

62.0

337.0

77.0

0.0

nbody

46.0

37.0

90.0

2.0

2.0

44.0

23.0

30.0

9.0

37.0

39.0

16.0

14.0

59.0

52.0

1598.0

108.0

0.0

nettle-aes

38.0

33.0

90.0

10.0

27.0

39.0

51.0

53.0

6.0

67.0

29.0

28.0

17.0

58.0

47.0

71.0

93.0

1.0

nettle-sha256

40.0

37.0

93.0

10.0

39.0

41.0

33.0

23.0

5.0

33.0

40.0

22.0

16.0

54.0

49.0

62.0

112.0

2.0

nsichneu

39.0

41.0

67.0

35.0

25.0

38.0

33.0

24.0

17.0

36.0

18.0

18.0

30.0

33.0

23.0

111.0

34.0

3.0

picojpeg

37.0

33.0

94.0

13.0

30.0

37.0

57.0

55.0

11.0

72.0

25.0

23.0

15.0

68.0

67.0

94.0

71.0

2.0

primecount

60.0

54.0

93.0

15.0

18.0

57.0

80.0

62.0

16.0

89.0

43.0

48.0

29.0

95.0

89.0

127.0

81.0

1.0

qrduino

54.0

51.0

115.0

18.0

26.0

56.0

84.0

68.0

11.0

95.0

39.0

37.0

25.0

110.0

110.0

113.0

82.0

2.0

sglib-combined

45.0

42.0

101.0

18.0

26.0

45.0

62.0

51.0

21.0

74.0

26.0

32.0

27.0

80.0

76.0

98.0

68.0

2.0

slre

55.0

49.0

131.0

35.0

46.0

56.0

71.0

63.0

22.0

93.0

34.0

27.0

31.0

104.0

69.0

139.0

87.0

3.0

st

42.0

34.0

86.0

2.0

2.0

41.0

19.0

25.0

8.0

31.0

35.0

13.0

14.0

59.0

57.0

907.0

140.0

0.0

statemate

52.0

50.0

181.0

39.0

56.0

51.0

66.0

44.0

24.0

66.0

32.0

31.0

34.0

141.0

109.0

203.0

148.0

3.0

tarfind

38.0

39.0

143.0

4.0

11.0

40.0

48.0

50.0

52.0

73.0

51.0

34.0

25.0

91.0

74.0

77.0

72.0

1.0

ud

36.0

32.0

79.0

8.0

15.0

36.0

34.0

38.0

16.0

47.0

29.0

18.0

15.0

51.0

46.0

73.0

43.0

1.0

wikisort

65.0

57.0

166.0

3.0

5.0

67.0

41.0

52.0

28.0

65.0

44.0

33.0

26.0

101.0

91.0

242.0

97.0

0.0

Notes:

  • This results present CPU performance compared to ARM Cortex M4

  • Higher values are better

Table 2 Benchmark statistics:

cv32e40p_standard_32_False

cv32e41p_standard_32_False

cva5_standard_32_False

femtorv_standard_32_False

firev_standard_32_False

ibex_standard_32_False

lm32_standard_32_False

marocchino_standard_32_False

minerva_standard_32_False

mor1kx_standard_32_False

naxriscv_standard_32_False

neorv32_standard_32_False

picorv32_standard_32_False

vexriscv_standard_32_False

vexriscv_smp_standard_32_False

blackparrot_sim_32_False

microwatt_standard+ghdl_32_False

serv_standard_32_False

speed geometric mean

44.0

39.0

99.0

8.0

12.0

43.0

47.0

44.0

15.0

58.0

34.0

26.0

19.0

74.0

65.0

147.0

81.0

1.0

speed geometric standard deviation

1.24

1.26

1.35

3.28

3.37

1.26

1.58

1.45

1.99

1.48

1.54

1.5

1.5

1.5

1.52

2.2

1.47

3.45

Notes:

  • Benchmark consists of multiple invidual tests

  • Higher geometric mean is better

  • Lower standard deviation is better, more consistent results

Table 3 Absolute tests results: [ms]

cv32e40p_standard_32_False

cv32e41p_standard_32_False

cva5_standard_32_False

femtorv_standard_32_False

firev_standard_32_False

ibex_standard_32_False

lm32_standard_32_False

marocchino_standard_32_False

minerva_standard_32_False

mor1kx_standard_32_False

naxriscv_standard_32_False

neorv32_standard_32_False

picorv32_standard_32_False

vexriscv_standard_32_False

vexriscv_smp_standard_32_False

blackparrot_sim_32_False

microwatt_standard+ghdl_32_False

serv_standard_32_False

aha-mont64

8623

9852

4881

110764

60125

8626

8963

7711

55335

7280

3417

13072

29142

5442

5618

3060

2581

1120897

crc32

9753

11146

4354

36912

16020

10623

6928

7319

42456

5750

9927

13757

26642

5399

5573

4355

5923

322476

cubic

16158

18209

8601

261289

172731

16549

20427

16710

61741

13570

26863

39381

47694

15056

17088

1648

3983

3058139

edn

9506

10812

4475

310023

225162

10573

6430

7357

9080

6706

16956

12925

39252

4583

5200

4463

7659

4039485

huffbench

8995

9432

3571

13155

9303

8830

5043

6717

19632

4695

11802

10535

14689

3669

4294

3457

5437

153910

matmult-int

10278

11758

4406

152955

116311

10866

7716

10119

6598

7112

19199

12224

30703

4889

6475

4382

8993

2015765

md5sum

6205

7089

2619

12806

7210

5968

4759

5245

21073

3734

6227

7982

12558

3194

3454

2916

3469

123217

minver

10614

12655

4767

323092

234827

10922

14797

14019

47807

11695

14660

26542

29792

6642

6434

1188

5191

4052253

nbody

6170

7617

3124

185188

130616

6399

12155

9487

31887

7693

7143

17046

19418

4722

5427

176

2610

2292102

nettle-aes

10604

12093

4482

39384

14797

10439

7878

7530

64986

5975

13817

14295

23986

6903

8637

5660

4316

316406

nettle-sha256

9907

10736

4286

38354

10361

9750

12190

17617

76438

12043

10085

18240

25545

7429

8142

6421

3568

226615

nsichneu

10392

9757

5940

11407

15978

10525

12021

16621

23328

11126

22379

22748

13194

12258

17581

3602

11674

157984

picojpeg

10929

12068

4279

31646

13521

11013

7113

7271

38344

5603

16085

17269

26024

5937

5995

4280

5674

257783

primecount

6386

7134

4108

25578

21480

6695

4801

6197

23438

4287

8872

7953

13325

4023

4285

3020

4758

337426

qrduino

7946

8289

3698

23428

16311

7596

5070

6221

40159

4492

10795

11628

17301

3882

3870

3757

5201

278813

sglib-combined

8773

9576

3938

22130

15495

8910

6397

7880

19050

5413

15414

12296

14879

4981

5245

4048

5858

245880

slre

7314

8190

3068

11437

8689

7207

5610

6333

18287

4307

11749

14954

12808

3851

5780

2892

4624

138148

st

9647

12016

4722

269502

188816

9876

21397

16474

50949

13122

11523

31186

29974

6884

7211

450

2919

3297097

statemate

7666

7946

2214

10130

7188

7791

6099

9027

16496

6094

12664

12945

11688

2844

3673

1966

2697

116687

tarfind

10630

10251

2812

101646

36955

9970

8486

8147

7734

5532

7846

11902

16028

4441

5476

5259

5586

761934

ud

11144

12673

5040

48754

26350

11223

11914

10469

25647

8421

13630

22423

27210

7820

8737

5458

9387

496181

wikisort

4287

4860

1673

83652

58508

4165

6738

5375

9951

4267

6356

8478

10786

2741

3047

1148

2879

1033451

Notes:

  • Lower value is better

Table 4 Benchmark statistics:

cv32e40p_standard_32_False

cv32e41p_standard_32_False

cva5_standard_32_False

femtorv_standard_32_False

firev_standard_32_False

ibex_standard_32_False

lm32_standard_32_False

marocchino_standard_32_False

minerva_standard_32_False

mor1kx_standard_32_False

naxriscv_standard_32_False

neorv32_standard_32_False

picorv32_standard_32_False

vexriscv_standard_32_False

vexriscv_smp_standard_32_False

blackparrot_sim_32_False

microwatt_standard+ghdl_32_False

serv_standard_32_False

speed geometric mean

8866

9839

3918

51606

31738

8957

8316

8857

25980

6674

11451

14932

20504

5270

6005

2645

4776

553415

speed geometric standard deviation

1.31

1.31

1.4

3.18

3.25

1.32

1.55

1.45

2.01

1.48

1.58

1.51

1.52

1.52

1.54

2.35

1.52

3.33

Notes:

  • Benchmark consists of multiple invidual tests

  • Lower geometric mean is better

  • Lower standard deviation is better, more consistent results

Platform

Core

SHA1

cv32e40p

c897dc0a291185a7ed6ef6904da1dc698909d6ed

cv32e41p

a48ddd9fc6f1cabfb3a7ac679774d4eff1f296fc

cva5

ea1075d60c82191a1391ad00aa55be916bb37ab7

ibex

d3d53df613c2878152eaa382a8cd558efdf21ec6

lm32

594f2068e32a6faa84f05f9b443b58c3c4658113

marocchino

ae64a18899f40607569833c8ad415c768d7fb3dc

minerva

fb0a0423ce861dfa75992551ca09401cc8c99b95

mor1kx

92628ec323dcce464185ad9e0f58a788c2be6989

naxriscv

8b808becb50383a4d213508016f9e8bfe28f5bc7

picorv32

a6e8fd113527b39c1d18837968de7e6b9b59ea92

vexriscv

e75700dff2ab9662f3e26dd89ab59a5f6da65687

e75700dff2ab9662f3e26dd89ab59a5f6da65687

blackparrot

ba50883f12d33e1d834640640c84ddc9329bb68a

microwatt

b940b55acffeb1ae791f2e083145bfb32b26eef9

serv

cdede3b5e35e0c8c385ff10d783905d95cd864ab

Used tools:

Toolchains:

RISC-V:

riscv64-elf-newlib-gcc (12.1.0) 12.1.0

LM32:

lm32-elf-gcc (9.2.0) 9.2.0

OpenRISC:

or1k-elf-gcc (9.2.0) 9.2.0

OpenPOWER:

ppc64le-linux-musl-gcc (12.1.0) 12.1.0

Other tools:

verilator:

Verilator 4.211 devel rev UNKNOWN.REV

yosys:

Yosys 0.20+45 (git sha1 029c2785e, x86_64-conda-linux-gnu-cc 11.2.0 -fvisibility-inlines-hidden -fmessage-length=0 -march=nocona -mtune=haswell -ftree-vectorize -fPIC -fstack-protector-strong -fno-plt -O2 -ffunction-sections -fdebug-prefix-map=/home/runner/work/conda-eda/conda-eda/workdir/conda-env/conda-bld/yosys_1660932005914/work=/usr/local/src/conda/yosys-0.20_46_g029c2785e -fdebug-prefix-map=/home/runner/work/embench-tester/embench-tester/env/conda/envs/embench-tester=/usr/local/src/conda-prefix -fPIC -Os -fno-merge-constants)

ghdl:

GHDL 2.0.0-dev (tarball) [Dunoon edition] Compiled with GNAT Version: 9.4.0 mcode code generator Written by Tristan Gingold.