Benchmarks 2024 11 26 TVM LLVM O3 spike_rv64 - tum-ei-eda/muriscv-nn GitHub Wiki
Setup
Simulator
- Spike (
riscv-isa-sim) (ISS, CPI=1)- Spike :
eb0a3e2b0a7c57522928be39de95cd9f8c6dc636 - Spike PK :
fix-gcc14-rvv
- Spike :
Toolchains
-
RISC-V GCC:
- Scalar:
riscv32-unknown-elf-gcc (g8b4bb54e6c4) 14.2.1 20241118 - Vector:
riscv32-unknown-elf-gcc (g8b4bb54e6c4) 14.2.1 20241118 - Packed: Self compiled using patches found in https://github.com/riscv-collab/riscv-gcc/pull/258 and https://github.com/riscvarchive/riscv-binutils-gdb/pull/257
- Scalar:
-
LLVM/Clang:
clang version 18.1.8 (https://github.com/llvm/llvm-project.git 3b5b5c1ec4a3095ab096dd780e84d7ab81f3d7ff)- Linker: lld (TODO)
Models
-
MLPerfTiny Benchmark
-
TODO: others!
Frameworks
-
MLonMCU :
develop -
TVM :
Nightly Pre-Build
Miscellaneous
- Used
-Osflag for compilation. - Benchmarks generated using MLonMCU deployment tool with minimal efforts.
- Memory metrics are reported in Bytes
Results (Framework: tvm, Backend: tvmaot, Toolchain: llvm, Flags: -O3, Target: spike_rv64 )
Audio Wake Words (aww)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
|---|---|---|---|---|---|---|---|---|---|
13151114.0 ( 1.2x ) |
125684 ( 1.432 ) |
61296 ( 2.911 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
20156761.0 ( 0.8x ) |
109016 ( 1.242 ) |
61288 ( 2.911 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
25193781.0 ( 0.6x ) |
179788 ( 2.049 ) |
61288 ( 2.911 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25140137.0 ( 0.6x ) |
179446 ( 2.045 ) |
61288 ( 2.911 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25283215.0 ( 0.6x ) |
180044 ( 2.051 ) |
61288 ( 2.911 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
26530169.0 ( 0.6x ) |
183682 ( 2.093 ) |
61288 ( 2.911 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
26536661.0 ( 0.6x ) |
183290 ( 2.088 ) |
61288 ( 2.911 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
26391069.0 ( 0.6x ) |
182978 ( 2.085 ) |
61288 ( 2.911 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
3633274.0 ( 4.3x ) |
127816 ( 1.456 ) |
61288 ( 2.911 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
2205916.0 ( 7.1x ) |
127868 ( 1.457 ) |
61288 ( 2.911 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
1470726.0 ( 10.7x ) |
125856 ( 1.434 ) |
61288 ( 2.911 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
725620.0 ( 21.6x ) |
115344 ( 1.314 ) |
61288 ( 2.911 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
736462.0 ( 21.3x ) |
115472 ( 1.316 ) |
61288 ( 2.911 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
736375.0 ( 21.3x ) |
115412 ( 1.315 ) |
61288 ( 2.911 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
15671855.0 ( Base ) |
87764 ( Base ) |
21056 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
15591560.0 ( 1.0x ) |
87656 ( 0.999 ) |
25520 ( 1.212 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
7073397.0 ( 2.2x ) |
96374 ( 1.098 ) |
21056 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6491057.0 ( 2.4x ) |
95912 ( 1.093 ) |
21056 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6200930.0 ( 2.5x ) |
95916 ( 1.093 ) |
21056 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6056799.0 ( 2.6x ) |
95916 ( 1.093 ) |
21056 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
14529412.0 ( 1.1x ) |
95916 ( 1.093 ) |
21056 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
14531277.0 ( 1.1x ) |
95916 ( 1.093 ) |
21056 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6005421.0 ( 2.6x ) |
90350 ( 1.029 ) |
25520 ( 1.212 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4871412.0 ( 3.2x ) |
89924 ( 1.025 ) |
25520 ( 1.212 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4313140.0 ( 3.6x ) |
89924 ( 1.025 ) |
25520 ( 1.212 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4274021.0 ( 3.7x ) |
89924 ( 1.025 ) |
25520 ( 1.212 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4274021.0 ( 3.7x ) |
89924 ( 1.025 ) |
25520 ( 1.212 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4275886.0 ( 3.7x ) |
89924 ( 1.025 ) |
25520 ( 1.212 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8199037.0 ( 1.9x ) |
97030 ( 1.106 ) |
25520 ( 1.212 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
7611578.0 ( 2.1x ) |
96608 ( 1.101 ) |
25520 ( 1.212 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
7321451.0 ( 2.1x ) |
96620 ( 1.101 ) |
25520 ( 1.212 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
7177320.0 ( 2.2x ) |
96620 ( 1.101 ) |
25520 ( 1.212 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15648941.0 ( 1.0x ) |
96572 ( 1.1 ) |
25520 ( 1.212 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15650806.0 ( 1.0x ) |
96572 ( 1.1 ) |
25520 ( 1.212 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
Image Classification (resnet)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
|---|---|---|---|---|---|---|---|---|---|
58410743.0 ( 1.0x ) |
226572 ( 1.677 ) |
110216 ( 1.921 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
67859894.0 ( 0.8x ) |
224162 ( 1.659 ) |
110216 ( 1.921 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
110154013.0 ( 0.5x ) |
293516 ( 2.172 ) |
110216 ( 1.921 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
109051500.0 ( 0.5x ) |
290948 ( 2.153 ) |
110216 ( 1.921 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
109583101.0 ( 0.5x ) |
291334 ( 2.156 ) |
110216 ( 1.921 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
109685479.0 ( 0.5x ) |
290348 ( 2.149 ) |
110216 ( 1.921 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
109915937.0 ( 0.5x ) |
289688 ( 2.144 ) |
110216 ( 1.921 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
110749310.0 ( 0.5x ) |
289922 ( 2.146 ) |
110216 ( 1.921 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
33941619.0 ( 1.6x ) |
221966 ( 1.643 ) |
110216 ( 1.921 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
30800702.0 ( 1.8x ) |
221988 ( 1.643 ) |
110216 ( 1.921 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
28432722.0 ( 2.0x ) |
220310 ( 1.631 ) |
110216 ( 1.921 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
27154118.0 ( 2.0x ) |
219536 ( 1.625 ) |
110216 ( 1.921 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
26663664.0 ( 2.1x ) |
218750 ( 1.619 ) |
110216 ( 1.921 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
26575287.0 ( 2.1x ) |
218184 ( 1.615 ) |
110216 ( 1.921 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
55600773.0 ( Base ) |
135116 ( Base ) |
57376 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
71380580.0 ( 0.8x ) |
134784 ( 0.998 ) |
57376 ( 1.0 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
13524487.0 ( 4.1x ) |
144616 ( 1.07 ) |
57376 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
11088249.0 ( 5.0x ) |
144102 ( 1.067 ) |
57376 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
13926693.0 ( 4.0x ) |
144102 ( 1.067 ) |
57376 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
15406836.0 ( 3.6x ) |
144102 ( 1.067 ) |
57376 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
16079408.0 ( 3.5x ) |
144102 ( 1.067 ) |
57376 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
35273375.0 ( 1.6x ) |
144102 ( 1.067 ) |
57376 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
15151856.0 ( 3.7x ) |
137842 ( 1.02 ) |
57376 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
9568658.0 ( 5.8x ) |
137324 ( 1.016 ) |
57376 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
7022818.0 ( 7.9x ) |
137324 ( 1.016 ) |
57376 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
5780690.0 ( 9.6x ) |
137324 ( 1.016 ) |
57376 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4854291.0 ( 11.5x ) |
137324 ( 1.016 ) |
57376 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4607064.0 ( 12.1x ) |
137324 ( 1.016 ) |
57376 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
16929120.0 ( 3.3x ) |
143938 ( 1.065 ) |
57376 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15894458.0 ( 3.5x ) |
143424 ( 1.061 ) |
57376 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
16646967.0 ( 3.3x ) |
143428 ( 1.062 ) |
57376 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
16355482.0 ( 3.4x ) |
143428 ( 1.062 ) |
57376 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15883222.0 ( 3.5x ) |
143428 ( 1.062 ) |
57376 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
42667589.0 ( 1.3x ) |
143428 ( 1.062 ) |
57376 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
Anomaly Detection (toycar)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
|---|---|---|---|---|---|---|---|---|---|
1629208.0 ( 1.0x ) |
582990 ( 1.863 ) |
7408 ( 1.121 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
1629208.0 ( 1.0x ) |
582990 ( 1.863 ) |
7408 ( 1.121 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
827111.0 ( 2.0x ) |
586314 ( 1.874 ) |
7408 ( 1.121 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
767987.0 ( 2.2x ) |
586068 ( 1.873 ) |
7408 ( 1.121 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
942838.0 ( 1.8x ) |
586912 ( 1.876 ) |
7408 ( 1.121 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
827944.0 ( 2.0x ) |
586822 ( 1.876 ) |
7408 ( 1.121 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
793852.0 ( 2.1x ) |
595760 ( 1.904 ) |
7408 ( 1.121 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
737143.0 ( 2.3x ) |
587926 ( 1.879 ) |
7408 ( 1.121 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
827111.0 ( 2.0x ) |
586314 ( 1.874 ) |
7408 ( 1.121 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
767987.0 ( 2.2x ) |
586068 ( 1.873 ) |
7408 ( 1.121 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
942838.0 ( 1.8x ) |
586912 ( 1.876 ) |
7408 ( 1.121 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
827944.0 ( 2.0x ) |
586822 ( 1.876 ) |
7408 ( 1.121 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
793852.0 ( 2.1x ) |
595760 ( 1.904 ) |
7408 ( 1.121 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
737143.0 ( 2.3x ) |
587926 ( 1.879 ) |
7408 ( 1.121 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
1658945.0 ( Base ) |
312866 ( Base ) |
6608 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
1658945.0 ( 1.0x ) |
312866 ( 1.0 ) |
6608 ( 1.0 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
582403.0 ( 2.8x ) |
314130 ( 1.004 ) |
6608 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
490149.0 ( 3.4x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
444933.0 ( 3.7x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
422325.0 ( 3.9x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
411021.0 ( 4.0x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
1349550.0 ( 1.2x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
623306.0 ( 2.7x ) |
314288 ( 1.005 ) |
6608 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
499787.0 ( 3.3x ) |
313874 ( 1.003 ) |
6608 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
440603.0 ( 3.8x ) |
313874 ( 1.003 ) |
6608 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
411347.0 ( 4.0x ) |
313874 ( 1.003 ) |
6608 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
407663.0 ( 4.1x ) |
313874 ( 1.003 ) |
6608 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
405779.0 ( 4.1x ) |
313874 ( 1.003 ) |
6608 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
582403.0 ( 2.8x ) |
314130 ( 1.004 ) |
6608 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
490149.0 ( 3.4x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
444933.0 ( 3.7x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
422325.0 ( 3.9x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
411021.0 ( 4.0x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
1349550.0 ( 1.2x ) |
313680 ( 1.003 ) |
6608 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
Visual Wake Words (vww)
| Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
|---|---|---|---|---|---|---|---|---|---|
37389018.0 ( 1.2x ) |
575098 ( 1.793 ) |
182880 ( 2.09 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
53415162.0 ( 0.8x ) |
551138 ( 1.718 ) |
182824 ( 2.089 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
67855139.0 ( 0.6x ) |
712782 ( 2.222 ) |
182832 ( 2.089 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
69111644.0 ( 0.6x ) |
717506 ( 2.237 ) |
182824 ( 2.089 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
68011322.0 ( 0.6x ) |
718656 ( 2.241 ) |
182824 ( 2.089 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
68681188.0 ( 0.6x ) |
721502 ( 2.25 ) |
182824 ( 2.089 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
68745809.0 ( 0.6x ) |
722470 ( 2.253 ) |
182824 ( 2.089 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
68508422.0 ( 0.6x ) |
721732 ( 2.25 ) |
182824 ( 2.089 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
11908833.0 ( 3.7x ) |
622782 ( 1.942 ) |
182824 ( 2.089 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
8008466.0 ( 5.5x ) |
621752 ( 1.939 ) |
182824 ( 2.089 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
5745353.0 ( 7.7x ) |
615936 ( 1.92 ) |
182824 ( 2.089 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
4738832.0 ( 9.3x ) |
603386 ( 1.881 ) |
182824 ( 2.089 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
4231032.0 ( 10.4x ) |
602658 ( 1.879 ) |
182824 ( 2.089 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
4196754.0 ( 10.5x ) |
606450 ( 1.891 ) |
182824 ( 2.089 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
43963094.0 ( Base ) |
320734 ( Base ) |
87512 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
44869073.0 ( 1.0x ) |
320760 ( 1.0 ) |
87512 ( 1.0 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
16952261.0 ( 2.6x ) |
330294 ( 1.03 ) |
87512 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
16531848.0 ( 2.7x ) |
328908 ( 1.025 ) |
87512 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
17711089.0 ( 2.5x ) |
328910 ( 1.025 ) |
87512 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
20862898.0 ( 2.1x ) |
328912 ( 1.025 ) |
87512 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
24388079.0 ( 1.8x ) |
328910 ( 1.025 ) |
87512 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
38352866.0 ( 1.1x ) |
328910 ( 1.025 ) |
87512 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
13374816.0 ( 3.3x ) |
324284 ( 1.011 ) |
87512 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
10048698.0 ( 4.4x ) |
323024 ( 1.007 ) |
87512 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8757279.0 ( 5.0x ) |
323026 ( 1.007 ) |
87512 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8255158.0 ( 5.3x ) |
323028 ( 1.007 ) |
87512 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8207392.0 ( 5.4x ) |
323028 ( 1.007 ) |
87512 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8209256.0 ( 5.4x ) |
323022 ( 1.007 ) |
87512 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
20591882.0 ( 2.1x ) |
330964 ( 1.032 ) |
87512 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
20319470.0 ( 2.2x ) |
329710 ( 1.028 ) |
87512 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
21837399.0 ( 2.0x ) |
329724 ( 1.028 ) |
87512 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
24989208.0 ( 1.8x ) |
329724 ( 1.028 ) |
87512 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
28506085.0 ( 1.5x ) |
329674 ( 1.028 ) |
87512 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
42470872.0 ( 1.0x ) |
329674 ( 1.028 ) |
87512 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
Original data
Click here to download the raw files for this benchmark.