1. I do have the latest version of magic 8.3.220 but the LVS issue persists. 2. I also tried opening the gds file in magic to view the drc errors i get but it does not display any blocks i am getting a blank screen(but i can see the gds in klayout!)

Can you include your lvs report?

only these exists

that is the output directory, not the temp directory

which is /tmp/openram_praveen_2114_temp

You showed me the contents of /home/praveen/OpenRAM/config_files/16x512_1KB_1/temp

Magic is quite slow at extracting the layout for LVS

klayout is multithreaded

here it is, getting the same LVS issue!

Ok so i tried to see some layouts in magic, and as u @User told i run the run_drc.sh , it generated some mag files, and i opened some and i had these kind of errors(images). So i was happy that it does make sense because in a youtube video with matt venn you just told that for magic DRC its gonna be only the metal layers which we check and i can see that! Here are my questions then, 1. So i got around 10 to 14 mag files(after running the sh file run_drc.sh) and i guess many of them have drc errors, so how should i actually approach with these, for getting a working sram ready? do i have to make these drc error free but after that how will the verilog and other files change according to my changes in the mag files if i did that.....it will really helpful if u can define a path for getting an sram block working after generating the files from the compiler ! Im really serious at generating them and will be happy to contribute by giving other sram macros(hopefully)! 2. And there are other sort of sh files like run_ext.sh., which i tried running and that throws a lot of mag file with the ext spice files for those also(so that also generated a sram_18_512_sky130.mag which magic just struggles to handle edits with 😞 ) but what are the other mag contribute to the core of this sram main mag file, because i can see sense amplifier, and precharge circuits mag files as one of them... 3. And what does the run_lvs.sh file do?

Subcircuit summary: Circuit 1: pinv |Circuit 2: pinv -------------------------------------------|------------------------------------------- sky130_fd_pr__nfet_01v8 (1) |sky130_fd_pr__nfet_01v8 (1) sky130_fd_pr__pfet_01v8 (1) |sky130_fd_pr__pfet_01v8 (1) Number of devices: 2 |Number of devices: 2 Number of nets: 5 Mismatch |Number of nets: 4 Mismatch --------------------------------------------------------------------------------------- NET mismatches: Class fragments follow (with fanout counts): Circuit 1: pinv |Circuit 2: pinv --------------------------------------------------------------------------------------- Net: A |Net: A sky130_fd_pr__pfet_01v8/2 = 1 | sky130_fd_pr__pfet_01v8/2 = 1 sky130_fd_pr__nfet_01v8/2 = 1 | sky130_fd_pr__nfet_01v8/2 = 1 | Net: Z |Net: Z sky130_fd_pr__nfet_01v8/(1|3) = 1 | sky130_fd_pr__pfet_01v8/(1|3) = 1 sky130_fd_pr__pfet_01v8/(1|3) = 1 | sky130_fd_pr__nfet_01v8/(1|3) = 1 | (no matching net) |Net: gnd | sky130_fd_pr__nfet_01v8/(1|3) = 1 | sky130_fd_pr__nfet_01v8/4 = 1 ---------------------------------------------------------------------------------------

Can you confirm which magic and netgen you are using?

(It's probably best to debug with a tiny design too..)

If you can run that and send me the entire temp directory it would be useful

here are the versions!

I think that's the issue.

Running the tiny example, will update shortly!

So is the temp meaning the output directory or the /tmp/ ?

So yes im getting the lvs and the magic error, so waiting for ur reply on the directory ques!

(or a random name. You can define this if you set OPENRAM_TMP, but then you can only run one copy at a time)

I don't know if .200 works either

meanwhile i downgrade my version of magic and check again

getting the same errors after changing the version too! will update you the /tmp/praveen_2329_temp files shortly

hope you have asker for this! https://github.com/PraveenRajAyyappan/openram_test

There are 16 DRC errors due to supply routing but those are somewhat expected and easy to fix

Do you get any errors when running run_ext.sh or run_drc.sh?

it says it is finished but that some errors and warning in between

The first error not the final

seems like there isnt any, but there are many errors with warning only

Work through it piece by piece...

The timestamp mismatch just means that the cells have different timestamps in the library

We overwrote the extracted cells with the library cells before running DRC

seems like the sky130 devices arent deteted?

no matching net

You can also look at the device output and see that the connection counts on the pins are different somewhere. If extra connections, that means there is a short. If there are missing connections, there is an open.

Tutorial 2