You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
…d by sky130 spare)
Problem was that there was the implicit assumption that
substitution cells would be top cells (or at least: not
child cells of other substitution cells).
I had not considered that case that a COMPONENT replacement cell would use other COMPONENT replacements as child cells. This is a more basic issue of the "create_multi_mapping" function.
I see your other point regarding the XOR results: one shape appears in the XOR output on layer 236/0 which is a false positive. That is a bug in the optimization we introduced: A cell is skipped when entirely covered by a box. That needs variant building as in one cell this is the case while in the other it isn't.
Test-case sent by mail.
The text was updated successfully, but these errors were encountered: