Do you have a platform enabled for it (similar to thenone in orfs)?

They do not have an ORFS platform enabled, no. We have a task to do so filed in our internal Jira.

@User thanks. Can you point me to the latest libs? Are they in open pdks

This is what I get:

$  find ./ | grep -i 'sky130_osu'
./OpenLane/pdks/open_pdks/sky130/openlane/sky130_osu_sc_t18
./OpenLane/pdks/open_pdks/sky130/openlane/sky130_osu_sc_t18/config.tcl
./OpenLane/pdks/open_pdks/sky130/openlane/sky130_osu_sc_t18/tracks.info
./OpenLane/pdks/sky130A/libs.tech/openlane/sky130_osu_sc_t18
./OpenLane/pdks/sky130A/libs.tech/openlane/sky130_osu_sc_t18/config.tcl
./OpenLane/pdks/sky130A/libs.tech/openlane/sky130_osu_sc_t18/tracks.info
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_12t_ls.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_12t_ms.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_12t_hs.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_15t_ls.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_15t_ms.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_15t_hs.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_18t_ls.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_18t_ms.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_18t_hs.sh
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_12t_ls.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_12t_ms.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_12t_hs.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_15t_ls.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_15t_ms.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_15t_hs.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_18t_ls.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_18t_ms.par
./OpenLane/pdks/sky130A/libs.tech/qflow/sky130_osu_sc_18t_hs.par

How do I generate the .lib .lef and .cdl files?

@User Sorry to keep you waiting. Give me 10 minutes and I will link you the files. They are in the Google repository structure.

Thanks!!

@User I'm back at the keyboard now. You can find the respective files at these locations: https://foss-eda-tools.googlesource.com/skywater-pdk/libs/sky130_osu_sc_t18/ https://foss-eda-tools.googlesource.com/skywater-pdk/libs/sky130_osu_sc_t15/ https://foss-eda-tools.googlesource.com/skywater-pdk/libs/sky130_osu_sc_t12/

The organization is as follow. Each repository contains cells of one particular size (# of tracks). These are divided into three variants: high-speed, medium-speed, and low-speed. The *`sky130_osu_sc_*T.tlef`* file at the top level is the tech-lef for a particular cell size (defines the placement grid). Within each variant's folder are

doc

,

gds

,

lef

,

lib

,

magic

,

spice

, and

verilog

directories. I thus sought to emulate the folder structure of other PDKs I had worked with.

Hooking them up to open_pdks, as well as setting up ORFS platforms, have been open action items on our internal Jira. I believe that they are important steps to make the cells more readily usable.

Thanks @User - I ll try these and get back to you. Just to clarify, there is no need for tapecell insertion for these libs since they are included in the cells

Correct

@User, @User: The OSU standard cell installations in open_pdks are the most lightly tested, so if anything appears not to work as advertised, please let me know.

@User thanks, based on my trials. There is an issue with the li1 routing. I think I can fet around it

I am using the 12t HS and I think the issue is fast or detail route. I will share the issue with Matt

Just a quick note @User, the OSU SC were designed to not allow for li1 routing. Their corresponding techlef's bottom routing layer should be m1, with li1 marked as

MASTERSLICE

I'm working on getting ORFS platforms up right now

the issue is fixed and already made a platform 🙂

we limited the routing to met1 and uncommented the li1 in tlef

we still need to make platforms for t18 and t15 though

so maybe you can make those but we should be able to quickly

I am out the rest of the day, I ll be back tomorrow

Roger. Do you have the 12T platform in a repository somewhere, so I can quickly edit it to 15T / 18T and try testing? If not, I'll continue on my current path

Regardless, enjoy your break!

I ll push it tonight - on the road

Drive safe

@User see https://github.com/tdene/OpenROAD-flow-scripts Ping me in the morning tomorrow, I can diff mine against yours and see if I missed anything. I'm setting up an ORFS environment and will start testing as well.

thanks I will take a look tomorrow

@User @User@User Hi, when I tried running DRC in klayout, it raised violations for the close nsdm shapes (red meshes) that are not merged. I found the nsdm layer to be 0.01um short compared to the site height (4.44um) . Also, do we have fill cells that have the nsdm and psdm layers? Thanks

@Anagha has left the channel