open-source-silicon.dev

1T1R.sch

1T1R.mag

1T1R_gds_drc_issue.png

1T1R.gds

I was seeing the following drc issues on extracting gds of a 1T1R cell <@U016EM8L91B>.  The issue does not show up in the magic layout prior to extraction. I am using the new volare sky130B pdk after installing from mpw-7 git repo.

```Via1 width &lt; 0.26um (via.1a + 2 * via.4a)
Metal2 overlap of Via1 &lt; 0.03um in one direction (met2.5 - met2.4)```

I don't know how you made that GDS file, but it has a library name that looks like a file path, which is causing issues with Magic.  I can modify Magic's code to avoid this, but the simpler solution is just to avoid naming the GDS library with a name that is not a legal GDS structure name, since the problem is that Magic wants to create a cell with the name of the GDS library.

Actually, the name of the cell itself has slashes in it, as well as the library name, and I don't know how that came about because both names are not legal syntax.

I updated Magic to automatically take care of this problem, although it does not solve the issue of how the cell name ended up with slashes in it in the first place.

I used `gds write` command on 1T1R.mag file I posted. 
Can this be a potential problem on running mpw7 pre checks as well?

.magicrc

Can there be something in the  .magicrc file I am using that's causing the problem?

The original magic file was named 1T1R.mag on which I did the gds write

Thanks you <@U016EM8L91B>, I tried the new magic install but I am still seeing the drc issue on an array 16x16 such cells, on setting the drc complete option

1T1R_16x16.sch

1T1R_16x16.gds

1T1R_16x16.mag

1T1R_16x16.sch,1T1R_16x16.gds,1T1R_16x16.mag

That's the standard `.magicrc` file.  I don't have any idea how the full path gets into the GDS file as the library and cell name.  It appears that the cell itself has the full path name, but since magic removes any path component when it reads a file (or at least it's supposed to, and always does for me), then there shouldn't be any way for that to happen.  Something is going wrong, either having to do with the file system, or the way you issue commands when you read the file, or something else;  but I have never seen that problem before and I don't know of any mechanism that would allow it to happen (obviously there is one, though).