gf180mcu standard cell spice library has diodes with unlabeled parameters

[d-m-bailey]

gf180mcuC/libs.ref/gf180mcu_fd_sc_mcu9t5v0/spice/gf180mcu_fd_sc_mcu9t5v0.spice has undefined values

.SUBCKT gf180mcu_fd_sc_mcu9t5v0__antenna I VDD VNW VPW VSS
D0 VPW I diode_nd2ps_06v0 0.2034p 1.85u M=1
D1 I VNW diode_pd2nw_06v0 0.2034p 1.85u M=1
.ENDS

whereas the extracted layout has

.subckt gf180mcu_fd_sc_mcu7t5v0__antenna VSS I VDD VNW VPW
D0 VPW I diode_nd2ps_06v0 pj=1.86u area=0.2052p
D1 I VNW diode_pd2nw_06v0 pj=1.86u area=0.2052p
.ends

[RTimothyEdwards]

Huh, the original CDL has that. Is that a CDL thing? Or an error on GF's part? If it's a CDL thing, then I need to write a fix into my script that converts CDL to SPICE. If it's a GF error, then it needs an issue in the issue tracker (on github at efabless/globalfoundries-pdk-libs-gf180mcu_fd_sc_mcu7t5v0).

[RTimothyEdwards]

@d-m-bailey : Do you have any idea if it's valid CDL syntax to have an area and perimeter of a diode without any parameter name in front of it? Feels like an error to me.

[RTimothyEdwards]

There is at least one other major error in the CDL, which is that the endcap cell is given four ports in the CDL. But endcap is just a copy of filltie, and has only two ports (VDD and VSS). So the existence of at least one other major error raises the probability that the diode issue is also just a straight-up error from GlobalFoundries.

[RTimothyEdwards]

Fixed in the last push to the repo on opencircuitdesign.com.