<@U028CCT3Q66> I don't believe that symbols have b...
# reram
t
@User I don't believe that symbols have been made, but #C017P3RAD42 might know better than I do
👍 1
t
There is a ReRAM symbol in xschem as of about 15 minutes ago.
❤️ 2
b
Hi @User , working with the xschem, particularly to access reram is not smooth, I feel. can you suggest a manual to begin reram circuit simulations. I have sky130 PDK on my institute server, able to perform non-reram simulations too. Trying to integrate the reram directory to access the symbol, without disturbing the installed PDK. The more specific manual will ease the initial setting-up.
t
Make sure you're using technology variant
sky130B
, which is the one that supports ReRAM. There are other issues, though, such as the fact that for ngspice, at least, the ReRAM verilog model would have to be compiled into ngspice itself. I have not tried it with Xyce, but in theory, it should be able to be compiled into a shared-object library that can be loaded into Xyce at run-time.
a
@Bhanprakash Goswami there are some xyce object compilation steps in the 1T1R example. https://github.com/google/skywater-pdk-libs-sky130_fd_pr_reram/tree/main/examples/1T1R I was able to use spack to install xyce but it took a while and needed Fortran compilers to be installed first. Haven’t quite run a sim yet but hopefully it’s not too many steps.
s
I am having trouble doing lvs as well. Reram devices show up in the xschem netlist and the netlist from magic. But lvs check does not recognise them as devices from xschem netlist
so lvs errors out saying number of devices do not match
netlists and netgen output for 4x4 crossbar array of 1T1R cell