Operating conditions in corners are wrong

[mguthaus]

Regarding liberty opcond, the problem is that some models use 1.9 in operating condition, but that is different to the voltage map, in order to support multi-voltage implementation, some tools rely on pg-pin voltage for linking, while others non MV-aware just assume single voltage by using operating condition alone. To avoid linking inconsistency, these should match in libraries modeling pg-pins.
sram_1rw1r_32_256_8_sky130_TT_1p9V_25C.lib

operating_conditions(OC){
process : 1.0 ;
voltage : 1.9 ;
temperature : 25;
}
...
nom_voltage : 1.8;
nom_temperature : 25;
nom_process : 1.0;
...
voltage_map ( VDD, 1.8 );
voltage_map ( GND, 0 );

[mguthaus]

Any update on this?

[hznichol]

I removed the nominal voltage/temp/process and made the voltage map the same as the operating conditions.
Was going to test at least the syntax before pushing the change.

[mguthaus]

Go ahead and push. You can assume the previous syntax was correct and just the values were wrong.

[mguthaus]

What is the status on this?

[hznichol]

Change has been added to which now has "operating_conditions" and "nom_voltage, nom_temperature, nom_process" as the same values in the lib file.