Can someone confirm using

drc euclidean

is correct for sky130 ? I'm doing something like : which passes using euclidean distance but fails manhattan distance ...

It works as far as I know. Do you have any specific issue that needs addressing?

I don't have an issue, the euclidean mode in magic seems to work. I just want to make sure it's OK to use it. I don't want to spend hours/days optimizing a layout and then be told skywater it's not manufacturable and I should have used manhattan distance only to check the various min.spacing rules.

The only thing weird about it is that it is implemented as a command an not specifically part of the tech file. It is not enabled by default in magic only to be backwards-compatible with use cases from something like 30 years ago. I have considered adding a tech file command to force it to be enabled when the tech file is read in. As it is currently done in sky130, the command is run as part of the startup script

sky130A.magicrc

, so if you start magic in the "recommended" way with

-rcfile /path/to/sky130A.magicrc

it will be in effect, but if you bypass the startup file and just run with

-T /path/to/sky130A.tech

then it won't, which is kind of stupid.