<@U016EM8L91B> I'm viewing a LEF/DEF which shows n...
# magic
m
@User I'm viewing a LEF/DEF which shows no DRC errors in the top of the window (green check), but there are error markings and when I highlight and do :drc why I get errors:
Copy code
Metal4 spacing < 0.3um (met4.2)
Metal4 > 3um spacing to unrelated m4 < 0.4um (met4.5b)
Metal5 width < 1.6um (met5.1)
Why does it show as passing at the top?
Some of the errors are just wrong too. The m5 width is 3um, but it complains that it is < 1.6um
t
Where the DRC errors are marked in the layout is over the contacts, so my guess is this is just related to the metal-over-obstruction problem; somewhere it is flattening the area for DRC checks, and since metal5 and obsm5 can't exist in the same area, probably it ended up with via4 contacts surrounded by obsm5, and only the via4 contacts are being counted for the m5 width. I could add obsm5 to the layers that are included in the metal5 width check, but I'm pretty sure I removed it on purpose because there are LEF views with thin strips of obsm5 that cause all sorts of DRC errors to show up if I don't.
👍 1