Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

LVS mismatch on parallel devices and issue on ambiguity resolution #1135

Closed
klayoutmatthias opened this issue Aug 2, 2022 · 0 comments
Closed
Assignees
Milestone

Comments

@klayoutmatthias
Copy link
Collaborator

klayoutmatthias commented Aug 2, 2022

Issue 1: mismatch on parallel devices

This may when device with different parameters are connected in parallel. The device comparison may report mismatching devices because the parallel ones are not associated correctly.

Here are two netlists that reproduce the problem:

.SUBCKT LVS_TEST 1 2
M1I38 1 2 1 1 PCH_18_MAC L=0.225U W=10653U AS=0P AD=0P PS=0U PD=0U
M1I39A 1 2 1 1 PCH_18_MAC L=0.135U W=1809U AS=0P AD=0P PS=0U PD=0U
.ENDS LVS_TEST

and

.SUBCKT lvs_test 1 2
M$1 1 2 1 1 PCH_18_MAC L=0.135U W=1809U
M$150 1 2 1 1 PCH_18_MAC L=0.225U W=10653U
.ENDS lvs_test

Issue 2: ambiguity resolution problem

This is the schematic of three inverter chains (2x5 inverters, 1x6 inverters):

.SUBCKT inv_chain 20 21
M$1 20 1 2 20 NCH_SVT_MAC L=0.016U W=3U 
M$2 20 2 3 20 NCH_SVT_MAC L=0.016U W=3U 
M$3 20 3 4 20 NCH_SVT_MAC L=0.016U W=3U 
M$4 20 4 5 20 NCH_SVT_MAC L=0.016U W=3U 
M$5 20 5 6 20 NCH_SVT_MAC L=0.016U W=3U 
M$6 20 6 7 20 NCH_SVT_MAC L=0.016U W=3U 
M$7 20 8 9 20 NCH_SVT_MAC L=0.016U W=3U 
M$8 20 9 10 20 NCH_SVT_MAC L=0.016U W=3U 
M$9 20 10 11 20 NCH_SVT_MAC L=0.016U W=3U 
M$10 20 11 12 20 NCH_SVT_MAC L=0.016U W=3U 
M$11 20 12 13 20 NCH_SVT_MAC L=0.016U W=3U 
M$12 20 14 15 20 NCH_SVT_MAC L=0.016U W=3U 
M$13 20 15 16 20 NCH_SVT_MAC L=0.016U W=3U 
M$14 20 16 17 20 NCH_SVT_MAC L=0.016U W=3U 
M$15 20 17 18 20 NCH_SVT_MAC L=0.016U W=3U 
M$16 20 18 19 20 NCH_SVT_MAC L=0.016U W=3U 
M$17 21 1 2 21 PCH_SVT_MAC L=0.016U W=3U 
M$18 21 2 3 21 PCH_SVT_MAC L=0.016U W=3U 
M$19 21 3 4 21 PCH_SVT_MAC L=0.016U W=3U 
M$20 21 4 5 21 PCH_SVT_MAC L=0.016U W=3U 
M$21 21 5 6 21 PCH_SVT_MAC L=0.016U W=3U 
M$22 21 6 7 21 PCH_SVT_MAC L=0.016U W=3U 
M$23 21 8 9 21 PCH_SVT_MAC L=0.016U W=3U 
M$24 21 9 10 21 PCH_SVT_MAC L=0.016U W=3U 
M$25 21 10 11 21 PCH_SVT_MAC L=0.016U W=3U 
M$26 21 11 12 21 PCH_SVT_MAC L=0.016U W=3U 
M$27 21 12 13 21 PCH_SVT_MAC L=0.016U W=3U 
M$28 21 14 15 21 PCH_SVT_MAC L=0.016U W=3U 
M$29 21 15 16 21 PCH_SVT_MAC L=0.016U W=3U 
M$30 21 16 17 21 PCH_SVT_MAC L=0.016U W=3U 
M$31 21 17 18 21 PCH_SVT_MAC L=0.016U W=3U 
M$32 21 18 19 21 PCH_SVT_MAC L=0.016U W=3U 
.ENDS inv_chain

Comparing such a netlist suffers from manyfold ambiguities and in the current version fails because the algorithm picks nodes that do not match. This is because currently ambiguity resolution is shortcut when the nodes lead to other nodes from the same ambiguity group. In this case the nodes between the inverters which form 13 topologically ambiguous nodes.

@klayoutmatthias klayoutmatthias self-assigned this Aug 2, 2022
@klayoutmatthias klayoutmatthias added this to the 0.27.11 milestone Aug 2, 2022
@klayoutmatthias klayoutmatthias changed the title LVS mismatch on parallel devices LVS mismatch on parallel devices and issue on ambiguity resolution Aug 6, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant