Skip to content

Fix double-wide OEB and the conb_1 LVS property mismatch - #120

Merged
jdicorpo merged 3 commits into
mainfrom
double-wide-production-ready
Oct 28, 2026
Merged

jdicorpo merged 3 commits into
mainfrom
double-wide-production-ready

Conversation

@marwaneltoukhy

Copy link
Copy Markdown
Contributor

Branch double-wide-production-ready from the double-wide production-ready work. See chipfoundry/double-wide-env PR 1 for the full set.

sky130_fd_sc_hd__conb_1's (and sky130_fd_sc_hvl__conb_1's) antenna-protection
res_generic_po resistor has a documented upstream spice-vs-layout mismatch:
the bundled spice model declares l=0.5, but the cell's actual drawn
geometry is l=0.045 (see google/skywater-pdk#276). Netgen's default 1%
length tolerance turns this stale-model artifact into a hard LVS property
mismatch ('Error: Property mismatch.') on an unmodified, standard PDK cell,
even when the circuits otherwise match uniquely.

res_generic_po is described by the tool's own author as 'more of a symbolic
layer used to separate nodes for LVS rather than an actual simulatable
device,' so dropping the length comparison for this one device (w stays
compared) is a safe, targeted fix -- it does not touch tolerance rules for
any other resistor device type (res_high_po, res_xhigh_po, etc.).
@jdicorpo
jdicorpo merged commit 9ed3121 into main Oct 28, 2026
8 checks passed
@jdicorpo
jdicorpo deleted the double-wide-production-ready branch October 28, 2026 15:37

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@marwaneltoukhy The discrepancy in the conb_1 size comes from a mismatch between the spice and gds files in the PDK.

The user is probably using an older version of the gds.

I would not recommend removing the length check because and outdated gds is being used.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants