The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl Design Languages >> Verilog-AMS >> segmentation fault when I simulate a verlioga opamp model https://designers-guide.org/forum/YaBB.pl?num=1248839423 Message started by hhd1983 on Jul 28th, 2009, 8:50pm |
Title: segmentation fault when I simulate a verlioga opamp model Post by hhd1983 on Jul 28th, 2009, 8:50pm Hi, I wrote a simple opamp model with veriloga, module test(in, ip,out); input in,ip; output out; electrical in, ip, out; parameter gain = 10000; parameter bw = 100; analog begin V(out) <+ laplace_nd(V(ip,in),{1},{1,1/(2*3.14*bw)}); end endmodule But the simulator tell me a segmentation fault when I simulated it with spectre. Can somebody help me? thanks. |
Title: Re: segmentation fault when I simulate a verlioga opamp model Post by Andrew Beckett on Jul 30th, 2009, 5:25am You'll need to give more than that. I just threw together a simple testbench, and it simulated OK in MMSIM71. So, you'll need to give the spectre netlist you're using, and ideally the subversion of spectre you're using (type "spectre -W"). Regards, andrew. |
Title: Re: segmentation fault when I simulate a verlioga opamp model Post by hhd1983 on Jul 30th, 2009, 7:21pm Thanks for the reply. The subversion of spectre is 5.10.41_USR3.102405, and the netlist is: Quote:
analysis type "tran" "dc" "ac" The file veriloga.va is: Quote:
That's all. Regards. |
Title: Re: segmentation fault when I simulate a verlioga opamp model Post by hhd1983 on Jul 30th, 2009, 7:25pm I find that it's no fault when I simulate it in Cadence 5.0.33. Is it a tool version problem? |
Title: Re: segmentation fault when I simulate a verlioga opamp model Post by Andrew Beckett on Aug 12th, 2009, 2:25pm Both the IC5141 version and 5.0.33 are old (really unsupported) versions of spectre. That said, I don't see the problem in the latest IC5141 version of spectre, nor with a more recent version (e.g. MMSIM71). That said, you didn't include the complete netlist including analysis statements, so it's hard to be sure. I just used a somewhat arbitrary transient analysis, but I don't know what you were doing. Also, your instance line for I7 uses a parameter "bandwith" (spelt wrong), whereas the veriloga has "bw". Not sure if that's relevant... I tried it both as "bandwith" and "bw" and it worked in both cases for me. So probably best just to try a more recent spectre version. Regards, Andrew. |
The Designer's Guide Community Forum » Powered by YaBB 2.2.2! YaBB © 2000-2008. All Rights Reserved. |