search

dgschwend / zynqnet

Master Thesis "ZynqNet: An FPGA-Accelerated Embedded Convolutional Neural Network"
GNU General Public License v3.0
743 stars 297 forks source link

"OPMODE Input Warning" Discussion #26

Open chengsonghust opened 6 years ago

chengsonghust commented 6 years ago

I have tried to fixed the "OPMODE Input Warning" . And this warning sames coming for "fpga_top_fadd_32ng8j.v" alone, which adds two floating data (32bit) using DSP48E. Other files including "fixed-point data ADD/MUL" and "float-point data MUL/CMP" do not induce this warning. This conclusion has verified after varying the data type and function of the Xilinx Tutorial "ug871-design-files/Using_IP_with_Zynq/lab1/hls_macc" .

void hls_macc(int a, int b, int *accum, bool accum_clr) { static acc_reg = 0; if (accum_clr) acc_reg = 0; *acc_reg += a b;* accum = acc_reg; }

It sames that using "fixed-point ADD" to replace "float-point ADD" is the only solution to avoid this warning. But it's really inelegant. And I still don't know the reason. Could any one give me more tips?

dgschwend commented 6 years ago

So this is still not fixed in the Xilinx libraries... It‘s just a warning, you can ignore it and the generated HDL code / FPGA logic still works. However, the warning slows down simulation so much that proper HDL simulation is not possible. Someone would need to file a bug report to Xilinx.

dgschwend commented 6 years ago

But anyway, using fixed-point instead of float would be a great idea for ZynqNet!