Y
y_p_w
Guest
Hi all-
Having a quick question that may be difficult to test in real
operation.
We are using an SDF to backannotate our netlist. We're looking to
override specific timing checks (setup/hold) to zero on specific
registers. We were manually going into the SDF and modifying them,
but we receive updates and were looking for a way to avoid the tedious
task of manually inserting the changes.
The tool used is Cadence Incisive Simulator (NC Affirma) version 6.1.
Someone suggested creating a separate SDF file, and using that to
override the setup and hold times for the list of cells by calling
this SDF file last. It seems to compile just fine, but I'm thinking
this could result in several possibilities:
1) The override works as intended - just on the specific SETUP/HOLD
parameters, and any previous annotated timing checks are intact.
2) The override works on the intended cells but all other timing
checks are purged. Doesn't seem likely.
3) It does override, but all other timing checks (that aren't
specified in the new timing definitions) revert back to the default
value in the standard cell.
Any idea what would happen? I guess the other thing we could do is
call our Cadence rep.
Having a quick question that may be difficult to test in real
operation.
We are using an SDF to backannotate our netlist. We're looking to
override specific timing checks (setup/hold) to zero on specific
registers. We were manually going into the SDF and modifying them,
but we receive updates and were looking for a way to avoid the tedious
task of manually inserting the changes.
The tool used is Cadence Incisive Simulator (NC Affirma) version 6.1.
Someone suggested creating a separate SDF file, and using that to
override the setup and hold times for the list of cells by calling
this SDF file last. It seems to compile just fine, but I'm thinking
this could result in several possibilities:
1) The override works as intended - just on the specific SETUP/HOLD
parameters, and any previous annotated timing checks are intact.
2) The override works on the intended cells but all other timing
checks are purged. Doesn't seem likely.
3) It does override, but all other timing checks (that aren't
specified in the new timing definitions) revert back to the default
value in the standard cell.
Any idea what would happen? I guess the other thing we could do is
call our Cadence rep.