Hi all,

I've run simulation speed tests for the 3 different flavors of GHDL
(mcode, llvm, gcc) and for 3 versions of the Xilinx Vivado simulator,
XSim (2015.1, 2015.2, 2015.3).

Launched on a x86_64 GNU/Linux system.

I used various applications generated with an HLS tool. The VHDL code
itself is not the point of interest, only simulation speed is.

Attached is the raw result file.

General trend:

- mcode is 2 times slower than gcc and llvm flavors at simulation, but
it launches immediately, which makes it ideal for small-scale
simulations.

- Flavors llvm and gcc present roughly the same perfs.

- Xilinx Vivado simulators are much slower than llvm and gcc flavors of
GHDL. It even seems that the newest version, 2015.3, is a huge
regression compared to version 2015.2.

Besides, for people who are interested in doing tests:
I run ArchLinux distibution and I prepared packages for the 3 GHDL
flavors. All are built from hg source.
mcode  https://aur.archlinux.org/packages/ghdl-mcode-hg/
llvm   https://aur.archlinux.org/packages/ghdl-llvm-hg/
gcc    https://aur.archlinux.org/packages/ghdl-hg/

Adrien
Execution time, in seconds

                   |   idct |  mjpeg |  adpcm |    aes |     bf |    gsm |   
mips | motion |    sha |  dfadd |  dfmul |  dfdiv |  dfsin ||  total |
===================================================================================================================================================
ghdl-mcode         |  0.600 |    313 |   39.5 |   4.13 |   68.5 |   3.52 |   
1.05 |   1.96 |   56.5 |  0.524 |  0.285 |   5.42 |    213 ||    708 |
===================================================================================================================================================
             analy |   1.47 |   10.1 |   8.04 |   3.52 |   2.78 |   4.15 |   
1.78 |   2.56 |   1.48 |   3.70 |   2.27 |   3.71 |   9.14 ||        |
ghdl-llvm    elabo |  0.190 |  0.638 |  0.436 |  0.252 |  0.248 |  0.317 |  
0.294 |  0.291 |  0.204 |  0.369 |  0.299 |  0.350 |  0.609 ||        |
               run |  0.182 |    118 |   14.2 |   1.50 |   27.4 |   1.23 |  
0.307 |  0.681 |   24.8 |  0.077 |  0.018 |   1.86 |   76.7 ||        |
             total |   1.84 |    129 |   22.7 |   5.27 |   30.4 |   5.70 |   
2.38 |   3.53 |   26.5 |   4.15 |   2.59 |   4.92 |   86.4 ||    325 |
===================================================================================================================================================
             analy |   1.82 |   8.59 |   7.51 |   3.94 |   3.21 |   4.60 |   
2.39 |   3.06 |   1.79 |   4.39 |   3.21 |   4.25 |   8.49 ||        |
ghdl-gcc     elabo |  0.164 |  0.530 |  0.359 |  0.219 |  0.202 |  0.274 |  
0.280 |  0.268 |  0.168 |  0.345 |  0.274 |  0.316 |  0.569 ||        |
               run |  0.176 |    123 |   14.1 |   1.56 |   27.3 |   1.22 |  
0.313 |  0.654 |   22.2 |  0.081 |  0.017 |   1.79 |   76.1 ||        |
             total |   2.16 |    132 |   22.0 |   5.72 |   30.7 |   6.09 |   
2.98 |   3.98 |   24.2 |   4.82 |   3.50 |   6.36 |   85.2 ||    330 |
===================================================================================================================================================

===================================================================================================================================================
             analy |  0.297 |  0.524 |  0.664 |  0.354 |  0.370 |  0.367 |  
0.325 |  0.345 |  0.322 |  0.515 |  0.479 |  0.578 |  0.769 ||        |
XSim 15.1    elabo |   7.61 |   22.1 |   12.8 |   6.68 |   5.01 |   6.25 |   
5.20 |   4.52 |   4.89 |   4.95 |   4.73 |   5.06 |   6.70 ||        |
               run |   4.20 |    203 |   34.7 |   6.05 |   42.8 |   5.74 |   
4.22 |   4.66 |   31.4 |   3.82 |   4.24 |   7.43 |    176 ||        |
             total |   12.1 |    226 |   48.2 |   13.1 |   48.2 |   12.4 |   
9.75 |   9.53 |   36.6 |   9.29 |   9.45 |   13.1 |    183 ||    631 |
===================================================================================================================================================
             analy |  0.298 |  0.545 |  0.630 |  0.347 |  0.373 |  0.374 |  
0.335 |  0.351 |  0.320 |  0.520 |  0.470 |  0.567 |  0.753 ||        |
XSim 15.2    elabo |   2.42 |   18.8 |   11.1 |   4.50 |   2.78 |   3.94 |   
2.98 |   2.38 |   2.68 |   2.74 |   2.53 |   2.88 |   4.67 ||        |
               run |   1.52 |    202 |   32.0 |   3.68 |   39.9 |   3.40 |   
1.83 |   2.33 |   29.0 |   1.47 |   1.34 |   5.01 |    172 ||        |
             total |   4.24 |    221 |   43.7 |   8.53 |   40.1 |   7.71 |   
5.15 |   5.06 |   32.0 |   4.73 |   4.34 |   8.46 |    177 ||    562 |
===================================================================================================================================================
             analy |  0.279 |  0.512 |  0.626 |  0.329 |  0.361 |  0.365 |  
0.316 |  0.355 |  0.302 |  0.503 |  0.471 |  0.557 |  0.768 ||        |
XSim 15.3    elabo |   2.04 |   19.1 |   11.2 |   4.49 |   2.41 |   4.27 |   
2.45 |   1.71 |   2.27 |   3.21 |   2.38 |   3.28 |   5.29 ||        |
               run |   1.66 |    386 |   56.6 |   5.85 |   76.3 |   5.18 |   
2.23 |   3.11 |   54.3 |   1.52 |   1.28 |   8.72 |    325 ||        |
             total |   3.98 |    406 |   68.4 |   10.7 |   79.1 |   9.82 |   
5.00 |   5.18 |   56.9 |   5.23 |   4.13 |   12.6 |    331 ||    998 |
===================================================================================================================================================
_______________________________________________
Ghdl-discuss mailing list
[email protected]
https://mail.gna.org/listinfo/ghdl-discuss

Reply via email to