1
+ {
2
+ "nbformat" : 4 ,
3
+ "nbformat_minor" : 0 ,
4
+ "metadata" : {
5
+ "colab" : {
6
+ "provenance" : [],
7
+ "gpuType" : " T4"
8
+ },
9
+ "kernelspec" : {
10
+ "name" : " python3" ,
11
+ "display_name" : " Python 3"
12
+ },
13
+ "language_info" : {
14
+ "name" : " python"
15
+ },
16
+ "accelerator" : " GPU"
17
+ },
18
+ "cells" : [
19
+ {
20
+ "cell_type" : " code" ,
21
+ "source" : [
22
+ " %%capture\n " ,
23
+ " \n " ,
24
+ " !git clone https://github.com/se-hwan/cusadi\n " ,
25
+ " !git clone -b feature/benchmarking https://github.com/based-robotics/jaxadi\n " ,
26
+ " # copy the content back to the current directory\n " ,
27
+ " !cp -r cusadi/** .\n " ,
28
+ " !pip install -e .\n " ,
29
+ " !cd jaxadi && pip install .\n " ,
30
+ " !cp jaxadi/benchmarks/run_benchmark.py ."
31
+ ],
32
+ "metadata" : {
33
+ "id" : " 6wCJ13jopz9v"
34
+ },
35
+ "execution_count" : 1 ,
36
+ "outputs" : []
37
+ },
38
+ {
39
+ "cell_type" : " code" ,
40
+ "source" : [
41
+ " from run_benchmark import main\n " ,
42
+ " main()"
43
+ ],
44
+ "metadata" : {
45
+ "colab" : {
46
+ "base_uri" : " https://localhost:8080/"
47
+ },
48
+ "id" : " UQRLUvYeHJxH" ,
49
+ "outputId" : " ab357d37-412c-4f86-ea26-c73e70901604"
50
+ },
51
+ "execution_count" : 2 ,
52
+ "outputs" : [
53
+ {
54
+ "output_type" : " stream" ,
55
+ "name" : " stdout" ,
56
+ "text" : [
57
+ " Generating CUDA code for CasADi function: fn_1e1\n " ,
58
+ " CUDA codegen complete for CasADi function: fn_1e1\n " ,
59
+ " Generating CUDA code for CasADi function: fn_1e2\n " ,
60
+ " CUDA codegen complete for CasADi function: fn_1e2\n " ,
61
+ " fn_1e1\n " ,
62
+ " fn_1e2\n " ,
63
+ " Running CUDA benchmark for 1 environments with function fn_1e1...\n " ,
64
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
65
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d4592d0>\n " ,
66
+ " Running CUDA benchmark for 2 environments with function fn_1e1...\n " ,
67
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
68
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459a80>\n " ,
69
+ " Running CUDA benchmark for 4 environments with function fn_1e1...\n " ,
70
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
71
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459870>\n " ,
72
+ " Running CUDA benchmark for 8 environments with function fn_1e1...\n " ,
73
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
74
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459ab0>\n " ,
75
+ " Running CUDA benchmark for 16 environments with function fn_1e1...\n " ,
76
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
77
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459e70>\n " ,
78
+ " Running CUDA benchmark for 32 environments with function fn_1e1...\n " ,
79
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
80
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459840>\n " ,
81
+ " Running CUDA benchmark for 64 environments with function fn_1e1...\n " ,
82
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
83
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459d80>\n " ,
84
+ " Running CUDA benchmark for 128 environments with function fn_1e1...\n " ,
85
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
86
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459960>\n " ,
87
+ " Running CUDA benchmark for 256 environments with function fn_1e1...\n " ,
88
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
89
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459930>\n " ,
90
+ " Running CUDA benchmark for 512 environments with function fn_1e1...\n " ,
91
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
92
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d4597b0>\n " ,
93
+ " Running CUDA benchmark for 1024 environments with function fn_1e1...\n " ,
94
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
95
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459990>\n " ,
96
+ " Running CUDA benchmark for 2048 environments with function fn_1e1...\n " ,
97
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
98
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459db0>\n " ,
99
+ " Running CUDA benchmark for 4096 environments with function fn_1e1...\n " ,
100
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
101
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459f30>\n " ,
102
+ " Running CUDA benchmark for 8192 environments with function fn_1e1...\n " ,
103
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
104
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d45a0b0>\n " ,
105
+ " Running CUDA benchmark for 16384 environments with function fn_1e1...\n " ,
106
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
107
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459810>\n " ,
108
+ " Running CUDA benchmark for 32768 environments with function fn_1e1...\n " ,
109
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
110
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d4599f0>\n " ,
111
+ " Running CUDA benchmark for 65536 environments with function fn_1e1...\n " ,
112
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
113
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d45a0e0>\n " ,
114
+ " Running CUDA benchmark for 131072 environments with function fn_1e1...\n " ,
115
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
116
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d459f90>\n " ,
117
+ " Running CUDA benchmark for 262144 environments with function fn_1e1...\n " ,
118
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
119
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d45a440>\n " ,
120
+ " Running CUDA benchmark for 524288 environments with function fn_1e1...\n " ,
121
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
122
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b3d45a290>\n " ,
123
+ " Running CUDA benchmark for 1048576 environments with function fn_1e1...\n " ,
124
+ " Loaded CasADi function: fn_1e1:(i0[4],i1[2],i2[2],i3)->(o0[2],o1[3x3],o2[3]) SXFunction\n " ,
125
+ " Loaded library: <CDLL '/content/build/libfn_1e1.so', handle 5bb692f4d500 at 0x7e0b406f7d00>\n " ,
126
+ " Running CUDA benchmark for 1 environments with function fn_1e2...\n " ,
127
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
128
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a3e0>\n " ,
129
+ " Running CUDA benchmark for 2 environments with function fn_1e2...\n " ,
130
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
131
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45ae90>\n " ,
132
+ " Running CUDA benchmark for 4 environments with function fn_1e2...\n " ,
133
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
134
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a800>\n " ,
135
+ " Running CUDA benchmark for 8 environments with function fn_1e2...\n " ,
136
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
137
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d459ae0>\n " ,
138
+ " Running CUDA benchmark for 16 environments with function fn_1e2...\n " ,
139
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
140
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a980>\n " ,
141
+ " Running CUDA benchmark for 32 environments with function fn_1e2...\n " ,
142
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
143
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a5f0>\n " ,
144
+ " Running CUDA benchmark for 64 environments with function fn_1e2...\n " ,
145
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
146
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45ada0>\n " ,
147
+ " Running CUDA benchmark for 128 environments with function fn_1e2...\n " ,
148
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
149
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a680>\n " ,
150
+ " Running CUDA benchmark for 256 environments with function fn_1e2...\n " ,
151
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
152
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45aef0>\n " ,
153
+ " Running CUDA benchmark for 512 environments with function fn_1e2...\n " ,
154
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
155
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a920>\n " ,
156
+ " Running CUDA benchmark for 1024 environments with function fn_1e2...\n " ,
157
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
158
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b160>\n " ,
159
+ " Running CUDA benchmark for 2048 environments with function fn_1e2...\n " ,
160
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
161
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b190>\n " ,
162
+ " Running CUDA benchmark for 4096 environments with function fn_1e2...\n " ,
163
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
164
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a890>\n " ,
165
+ " Running CUDA benchmark for 8192 environments with function fn_1e2...\n " ,
166
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
167
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b3d0>\n " ,
168
+ " Running CUDA benchmark for 16384 environments with function fn_1e2...\n " ,
169
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
170
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b010>\n " ,
171
+ " Running CUDA benchmark for 32768 environments with function fn_1e2...\n " ,
172
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
173
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b400>\n " ,
174
+ " Running CUDA benchmark for 65536 environments with function fn_1e2...\n " ,
175
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
176
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45a8c0>\n " ,
177
+ " Running CUDA benchmark for 131072 environments with function fn_1e2...\n " ,
178
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
179
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b640>\n " ,
180
+ " Running CUDA benchmark for 262144 environments with function fn_1e2...\n " ,
181
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
182
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b340>\n " ,
183
+ " Running CUDA benchmark for 524288 environments with function fn_1e2...\n " ,
184
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
185
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b700>\n " ,
186
+ " Running CUDA benchmark for 1048576 environments with function fn_1e2...\n " ,
187
+ " Loaded CasADi function: fn_1e2:(i0[9],i1[3],i2[6],i3[2])->(o0[3],o1[5x5,23nz],o2[5]) SXFunction\n " ,
188
+ " Loaded library: <CDLL '/content/build/libfn_1e2.so', handle 5bb69380ab70 at 0x7e0b3d45b520>\n " ,
189
+ " Running Jaxadi benchmark for 1 environments with function fn_1e1...\n " ,
190
+ " Running Jaxadi benchmark for 2 environments with function fn_1e1...\n " ,
191
+ " Running Jaxadi benchmark for 4 environments with function fn_1e1...\n " ,
192
+ " Running Jaxadi benchmark for 8 environments with function fn_1e1...\n " ,
193
+ " Running Jaxadi benchmark for 16 environments with function fn_1e1...\n " ,
194
+ " Running Jaxadi benchmark for 32 environments with function fn_1e1...\n " ,
195
+ " Running Jaxadi benchmark for 64 environments with function fn_1e1...\n " ,
196
+ " Running Jaxadi benchmark for 128 environments with function fn_1e1...\n " ,
197
+ " Running Jaxadi benchmark for 256 environments with function fn_1e1...\n " ,
198
+ " Running Jaxadi benchmark for 512 environments with function fn_1e1...\n " ,
199
+ " Running Jaxadi benchmark for 1024 environments with function fn_1e1...\n " ,
200
+ " Running Jaxadi benchmark for 2048 environments with function fn_1e1...\n " ,
201
+ " Running Jaxadi benchmark for 4096 environments with function fn_1e1...\n " ,
202
+ " Running Jaxadi benchmark for 8192 environments with function fn_1e1...\n " ,
203
+ " Running Jaxadi benchmark for 16384 environments with function fn_1e1...\n " ,
204
+ " Running Jaxadi benchmark for 32768 environments with function fn_1e1...\n " ,
205
+ " Running Jaxadi benchmark for 65536 environments with function fn_1e1...\n " ,
206
+ " Running Jaxadi benchmark for 131072 environments with function fn_1e1...\n " ,
207
+ " Running Jaxadi benchmark for 262144 environments with function fn_1e1...\n " ,
208
+ " Running Jaxadi benchmark for 524288 environments with function fn_1e1...\n " ,
209
+ " Running Jaxadi benchmark for 1048576 environments with function fn_1e1...\n " ,
210
+ " Running Jaxadi benchmark for 1 environments with function fn_1e2...\n " ,
211
+ " Running Jaxadi benchmark for 2 environments with function fn_1e2...\n " ,
212
+ " Running Jaxadi benchmark for 4 environments with function fn_1e2...\n " ,
213
+ " Running Jaxadi benchmark for 8 environments with function fn_1e2...\n " ,
214
+ " Running Jaxadi benchmark for 16 environments with function fn_1e2...\n " ,
215
+ " Running Jaxadi benchmark for 32 environments with function fn_1e2...\n " ,
216
+ " Running Jaxadi benchmark for 64 environments with function fn_1e2...\n " ,
217
+ " Running Jaxadi benchmark for 128 environments with function fn_1e2...\n " ,
218
+ " Running Jaxadi benchmark for 256 environments with function fn_1e2...\n " ,
219
+ " Running Jaxadi benchmark for 512 environments with function fn_1e2...\n " ,
220
+ " Running Jaxadi benchmark for 1024 environments with function fn_1e2...\n " ,
221
+ " Running Jaxadi benchmark for 2048 environments with function fn_1e2...\n " ,
222
+ " Running Jaxadi benchmark for 4096 environments with function fn_1e2...\n " ,
223
+ " Running Jaxadi benchmark for 8192 environments with function fn_1e2...\n " ,
224
+ " Running Jaxadi benchmark for 16384 environments with function fn_1e2...\n " ,
225
+ " Running Jaxadi benchmark for 32768 environments with function fn_1e2...\n " ,
226
+ " Running Jaxadi benchmark for 65536 environments with function fn_1e2...\n " ,
227
+ " Running Jaxadi benchmark for 131072 environments with function fn_1e2...\n " ,
228
+ " Running Jaxadi benchmark for 262144 environments with function fn_1e2...\n " ,
229
+ " Running Jaxadi benchmark for 524288 environments with function fn_1e2...\n " ,
230
+ " Running Jaxadi benchmark for 1048576 environments with function fn_1e2...\n " ,
231
+ " Benchmark results saved.\n "
232
+ ]
233
+ }
234
+ ]
235
+ }
236
+ ]
237
+ }
0 commit comments