#verification-be #openram @User I thought a remembered a recent change to magic to extract the dummy devices in the sram bitcells. This is what I get with magic 8.3.179 - just the odd sized devices. (I didn't see any relevant changes in the commit messages, so I haven't tried the latest version.)

.subckt pk_sky130_fd_bd_sram__openram_dp_cell_replica VDD BL0 BL1 BR0 BR1 GND WL0
+ WL1 a_400_n79# a_38_n79#
...
X0 a_38_133# WL0 a_38_133# VDD sky130_fd_pr__special_pfet_pass ... w=70000u l=150000u
X4 VDD WL1 VDD VDD sky130_fd_pr__special_pfet_pass ... w=70000u l=150000u

X3 GND GND BL1 GND sky130_fd_pr__special_nfet_latch ... w=210000u l=80000u
X6 GND GND a_400_n79# GND sky130_fd_pr__special_nfet_latch ... w=210000u l=80000u
X9 GND GND a_38_n79# GND sky130_fd_pr__special_nfet_latch ... w=210000u l=80000u
X10 GND GND BR1 GND sky130_fd_pr__special_nfet_latch ... w=210000u l=80000u
.ends

and this is what's in the spice netlist from https://github.com/efabless/sky130_sram_macros/blob/main/sram_1rw1r_32_256_8_sky130/sram_1rw1r_32_256_8_sky130.lvs.sp .

.SUBCKT sky130_fd_bd_sram__openram_dp_cell_replica bl0 br0 bl1 br1 wl0 wl1 vdd gnd
** N=9 EP=8 IP=0 FDC=16
*.SEEDPROM

...

* drainOnly PMOS
X10 vdd wl1 vdd vdd sky130_fd_pr__special_pfet_pass L=0.08 W=0.14 m=1 mult=1
X11 Q wl0 Q vdd sky130_fd_pr__special_pfet_pass L=0.08 W=0.14 m=1 mult=1

* drainOnly NMOS
X12 bl1 gnd bl1 gnd sky130_fd_pr__special_nfet_latch W=0.21 L=0.08 m=1 mult=1
X14 br1 gnd br1 gnd sky130_fd_pr__special_nfet_latch W=0.21 L=0.08 m=1 mult=1

.ENDS

You'll notice that the number and connections of the nfets differs along with the pmos parameters. @User Is it correct to change the spice netlist to match the extracted netlist or do we need the same result as some other tool, such as Calibre?

The parasitic devices are odd because they are not really complete transistors. I'm not sure what the "best" methodology should be. Calibre also did its own hacks by using labels to identify devices which seems error prone to me as well.