Benchmarks 2024 11 21 TVM LLVM O3 spike_rv64 - tum-ei-eda/muriscv-nn GitHub Wiki
Setup
Simulator
- Spike (
riscv-isa-sim
) (ISS, CPI=1)- Spike :
0bc176b3fca43560b9e8586cdbc41cfde073e17a
- Spike PK :
7e9b671c0415dfd7b562ac934feb9380075d4aa2
- Spike :
Toolchains
-
RISC-V GCC:
- Scalar:
riscv32-unknown-elf-gcc (gc891d8dc23e) 13.2.0
- Vector:
riscv32-unknown-elf-gcc (gc891d8dc23e) 13.2.0
- 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
-Os
flag 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 )
aww
)
Audio Wake Words (Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
---|---|---|---|---|---|---|---|---|---|
13151076 ( 1.2x ) |
123276 ( 1.444 ) |
61288 ( 2.912 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
20156731 ( 0.8x ) |
106608 ( 1.249 ) |
61280 ( 2.911 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
25188588 ( 0.6x ) |
177380 ( 2.078 ) |
61280 ( 2.911 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25190699 ( 0.6x ) |
177380 ( 2.078 ) |
61280 ( 2.911 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25193467 ( 0.6x ) |
177380 ( 2.078 ) |
61280 ( 2.911 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25199018 ( 0.6x ) |
177380 ( 2.078 ) |
61280 ( 2.911 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25211658 ( 0.6x ) |
177380 ( 2.078 ) |
61280 ( 2.911 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25235903 ( 0.6x ) |
177380 ( 2.078 ) |
61280 ( 2.911 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
3633236 ( 4.3x ) |
125408 ( 1.469 ) |
61280 ( 2.911 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
2200422 ( 7.1x ) |
125408 ( 1.469 ) |
61280 ( 2.911 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
1484795 ( 10.6x ) |
125408 ( 1.469 ) |
61280 ( 2.911 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
1637211 ( 9.6x ) |
125408 ( 1.469 ) |
61280 ( 2.911 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
20562293 ( 0.8x ) |
125408 ( 1.469 ) |
61280 ( 2.911 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
20564158 ( 0.8x ) |
125408 ( 1.469 ) |
61280 ( 2.911 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
15676964 ( Base ) |
85356 ( Base ) |
21048 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
15585850 ( 1.0x ) |
85248 ( 0.999 ) |
25512 ( 1.212 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
7078502 ( 2.2x ) |
93966 ( 1.101 ) |
21048 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6492170 ( 2.4x ) |
93966 ( 1.101 ) |
21048 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6202043 ( 2.5x ) |
93966 ( 1.101 ) |
21048 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6057912 ( 2.6x ) |
93966 ( 1.101 ) |
21048 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
14530525 ( 1.1x ) |
93966 ( 1.101 ) |
21048 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
14532390 ( 1.1x ) |
93966 ( 1.101 ) |
21048 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
6000272 ( 2.6x ) |
87942 ( 1.03 ) |
25512 ( 1.212 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4866264 ( 3.2x ) |
87942 ( 1.03 ) |
25512 ( 1.212 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4307992 ( 3.6x ) |
87942 ( 1.03 ) |
25512 ( 1.212 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4268873 ( 3.7x ) |
87942 ( 1.03 ) |
25512 ( 1.212 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4268873 ( 3.7x ) |
87942 ( 1.03 ) |
25512 ( 1.212 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4270738 ( 3.7x ) |
87942 ( 1.03 ) |
25512 ( 1.212 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8198450 ( 1.9x ) |
94596 ( 1.108 ) |
25512 ( 1.212 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
7612118 ( 2.1x ) |
94596 ( 1.108 ) |
25512 ( 1.212 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
7321991 ( 2.1x ) |
94596 ( 1.108 ) |
25512 ( 1.212 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
7177860 ( 2.2x ) |
94596 ( 1.108 ) |
25512 ( 1.212 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15650473 ( 1.0x ) |
94596 ( 1.108 ) |
25512 ( 1.212 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15654203 ( 1.0x ) |
94596 ( 1.108 ) |
25512 ( 1.212 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
resnet
)
Image Classification (Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
---|---|---|---|---|---|---|---|---|---|
58410657 ( 1.0x ) |
224164 ( 1.689 ) |
110208 ( 1.921 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
67859808 ( 0.8x ) |
221728 ( 1.671 ) |
110208 ( 1.921 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
110153895 ( 0.5x ) |
291112 ( 2.194 ) |
110208 ( 1.921 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
110118149 ( 0.5x ) |
291108 ( 2.194 ) |
110208 ( 1.921 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
110327550 ( 0.5x ) |
291108 ( 2.194 ) |
110208 ( 1.921 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
110333107 ( 0.5x ) |
291112 ( 2.194 ) |
110208 ( 1.921 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
110341764 ( 0.5x ) |
291112 ( 2.194 ) |
110208 ( 1.921 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
110360412 ( 0.5x ) |
291108 ( 2.194 ) |
110208 ( 1.921 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
33941538 ( 1.6x ) |
219572 ( 1.655 ) |
110208 ( 1.921 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
30816398 ( 1.8x ) |
219558 ( 1.654 ) |
110208 ( 1.921 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
29337411 ( 1.9x ) |
219560 ( 1.654 ) |
110208 ( 1.921 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
53821895 ( 1.0x ) |
219572 ( 1.655 ) |
110208 ( 1.921 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
70125313 ( 0.8x ) |
219558 ( 1.654 ) |
110208 ( 1.921 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
70127178 ( 0.8x ) |
219558 ( 1.654 ) |
110208 ( 1.921 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
55595602 ( Base ) |
132708 ( Base ) |
57368 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
71339536 ( 0.8x ) |
132376 ( 0.997 ) |
57368 ( 1.0 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
13519312 ( 4.1x ) |
142208 ( 1.072 ) |
57368 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
11083068 ( 5.0x ) |
142208 ( 1.072 ) |
57368 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
13889346 ( 4.0x ) |
142208 ( 1.072 ) |
57368 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
15369489 ( 3.6x ) |
142208 ( 1.072 ) |
57368 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
16042061 ( 3.5x ) |
142208 ( 1.072 ) |
57368 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
35236028 ( 1.6x ) |
142208 ( 1.072 ) |
57368 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
15156941 ( 3.7x ) |
135434 ( 1.021 ) |
57368 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
9573741 ( 5.8x ) |
135434 ( 1.021 ) |
57368 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
7027901 ( 7.9x ) |
135434 ( 1.021 ) |
57368 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
5785773 ( 9.6x ) |
135434 ( 1.021 ) |
57368 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4859374 ( 11.4x ) |
135434 ( 1.021 ) |
57368 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
4612147 ( 12.1x ) |
135434 ( 1.021 ) |
57368 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
16882943 ( 3.3x ) |
141530 ( 1.066 ) |
57368 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15848275 ( 3.5x ) |
141530 ( 1.066 ) |
57368 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
16600784 ( 3.3x ) |
141530 ( 1.066 ) |
57368 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
16309299 ( 3.4x ) |
141530 ( 1.066 ) |
57368 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
15837039 ( 3.5x ) |
141530 ( 1.066 ) |
57368 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
42621406 ( 1.3x ) |
141530 ( 1.066 ) |
57368 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
toycar
)
Anomaly Detection (Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
---|---|---|---|---|---|---|---|---|---|
1628946 ( 1.0x ) |
580582 ( 1.87 ) |
7400 ( 1.121 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
1628946 ( 1.0x ) |
580582 ( 1.87 ) |
7400 ( 1.121 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
821720 ( 2.0x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
763231 ( 2.2x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
742171 ( 2.2x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
729642 ( 2.3x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
722445 ( 2.3x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
862318 ( 1.9x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
821720 ( 2.0x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
763231 ( 2.2x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
742171 ( 2.2x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
729642 ( 2.3x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
722445 ( 2.3x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
862318 ( 1.9x ) |
583906 ( 1.881 ) |
7400 ( 1.121 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
1663942 ( Base ) |
310454 ( Base ) |
6600 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
1663942 ( 1.0x ) |
310454 ( 1.0 ) |
6600 ( 1.0 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
577144 ( 2.9x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
491116 ( 3.4x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
445900 ( 3.7x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
423292 ( 3.9x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
411988 ( 4.0x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
1350517 ( 1.2x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
618047 ( 2.7x ) |
311876 ( 1.005 ) |
6600 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
499679 ( 3.3x ) |
311876 ( 1.005 ) |
6600 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
440495 ( 3.8x ) |
311876 ( 1.005 ) |
6600 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
411239 ( 4.0x ) |
311876 ( 1.005 ) |
6600 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
407555 ( 4.1x ) |
311876 ( 1.005 ) |
6600 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
405671 ( 4.1x ) |
311876 ( 1.005 ) |
6600 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
577144 ( 2.9x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
491116 ( 3.4x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
445900 ( 3.7x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
423292 ( 3.9x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
411988 ( 4.0x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
1350517 ( 1.2x ) |
311718 ( 1.004 ) |
6600 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
vww
)
Visual Wake Words (Cycles (Speedup) | Total ROM (rel.) | Total RAM (rel.) | VLEN | Layout | Kernels | Mode | Arch | Unroll | Auto-Vectorization |
---|---|---|---|---|---|---|---|---|---|
37388763 ( 1.2x ) |
572690 ( 1.799 ) |
182872 ( 2.09 ) |
0 | NCHW | TVM | Fallback | RV64GC | 0 | - |
53420041 ( 0.8x ) |
548724 ( 1.724 ) |
182816 ( 2.089 ) |
0 | NHWC | TVM | Fallback | RV64GC | 0 | - |
67859943 ( 0.6x ) |
710370 ( 2.232 ) |
182824 ( 2.089 ) |
128 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
67815248 ( 0.6x ) |
710370 ( 2.232 ) |
182824 ( 2.089 ) |
256 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
67817113 ( 0.6x ) |
710370 ( 2.232 ) |
182824 ( 2.089 ) |
512 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
67822708 ( 0.6x ) |
710370 ( 2.232 ) |
182824 ( 2.089 ) |
1024 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
67833897 ( 0.6x ) |
710368 ( 2.232 ) |
182824 ( 2.089 ) |
2048 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
67856280 ( 0.6x ) |
710374 ( 2.232 ) |
182824 ( 2.089 ) |
4096 | NCHW | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
11903426 ( 3.7x ) |
620378 ( 1.949 ) |
182816 ( 2.089 ) |
128 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
8466268 ( 5.2x ) |
620380 ( 1.949 ) |
182816 ( 2.089 ) |
256 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
9189144 ( 4.8x ) |
620372 ( 1.949 ) |
182816 ( 2.089 ) |
512 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
17558774 ( 2.5x ) |
620372 ( 1.949 ) |
182816 ( 2.089 ) |
1024 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
25851857 ( 1.7x ) |
620380 ( 1.949 ) |
182816 ( 2.089 ) |
2048 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
50337939 ( 0.9x ) |
620378 ( 1.949 ) |
182816 ( 2.089 ) |
4096 | NHWC | TVM | Fallback | RV64GCV | 0 | Loop+SLP |
43962962 ( Base ) |
318326 ( Base ) |
87504 ( Base ) |
0 | NHWC | muRISCV-NN | Scalar | RV64GC | 0 | - |
44776163 ( 1.0x ) |
318352 ( 1.0 ) |
87504 ( 1.0 ) |
0 | NHWC | muRISCV-NN | Vector (Portable) | RV64GC | 0 | - |
16957252 ( 2.6x ) |
327884 ( 1.03 ) |
87504 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
16536868 ( 2.7x ) |
327886 ( 1.03 ) |
87504 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
17695660 ( 2.5x ) |
327886 ( 1.03 ) |
87504 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
20847469 ( 2.1x ) |
327886 ( 1.03 ) |
87504 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
24372650 ( 1.8x ) |
327886 ( 1.03 ) |
87504 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
38337437 ( 1.1x ) |
327884 ( 1.03 ) |
87504 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Scalar | RV64GCV | 0 | Loop+SLP |
13369553 ( 3.3x ) |
321874 ( 1.011 ) |
87504 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
10048568 ( 4.4x ) |
321872 ( 1.011 ) |
87504 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8757149 ( 5.0x ) |
321876 ( 1.011 ) |
87504 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8255028 ( 5.3x ) |
321874 ( 1.011 ) |
87504 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8207262 ( 5.4x ) |
321876 ( 1.011 ) |
87504 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
8209127 ( 5.4x ) |
321876 ( 1.011 ) |
87504 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector | RV64GCV | 0 | - |
20504097 ( 2.1x ) |
328556 ( 1.032 ) |
87504 ( 1.0 ) |
128 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
20226561 ( 2.2x ) |
328556 ( 1.032 ) |
87504 ( 1.0 ) |
256 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
21744490 ( 2.0x ) |
328556 ( 1.032 ) |
87504 ( 1.0 ) |
512 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
24896299 ( 1.8x ) |
328556 ( 1.032 ) |
87504 ( 1.0 ) |
1024 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
28421479 ( 1.5x ) |
328552 ( 1.032 ) |
87504 ( 1.0 ) |
2048 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
42388131 ( 1.0x ) |
328552 ( 1.032 ) |
87504 ( 1.0 ) |
4096 | NHWC | muRISCV-NN | Vector (Portable) | RV64GCV | 0 | Loop+SLP |
Original data
Click here to download the raw files for this benchmark.