usim_trim to change block parameter multiplicity (m)

usim_trim can only be used with a small set of parameters, as documented.

You could just use an alter instead:

trim1 alter dev=m1 param=m value=4

(the first word is a unique name to distinguish it from any other alter statements).

Regards,

Andrew.

Thanks Andrew,

it is still not clear for me what is the dev=m1

let's say my instance in netlist is :

XSEG\<7\>  (sl\<0\>  bl\<0\> 0 vdd)  SEG_RC1X 

I want to add a multiplicity parameter of m=4 without touching the netlist. can I use the alter option directly OR the instance in netlist must already have a parameter m:

(XSEG\<7\> (sl\<0\>  bl\<0\> 0 vdd) SEG_RC1X  m=m1) ?

trim1 alter param=m1 value=4

Thanks,

Regards,

Kotb

Yes, this still works. You do need to put:

trim1 alter dev=XSEG\<7\> param=m value=4

The dev is the instance name. You can change the m-factor on a hierarchical instance, and will then apply m to everything within it (multiplying any existing m-factors on those lower level blocks by the m-factor on the instance).

You can add:

outPar info what=output where=file

into the netlist, and then you can see the m-factor on each of the lower level effective parameters (it will write a file called netlistSuffix.info.output into the working dir with the information for the output parameters of each device).

Andrew.

Yes, this still works. You do need to put:

trim1 alter dev=XSEG\<7\> param=m value=4

The dev is the instance name. You can change the m-factor on a hierarchical instance, and will then apply m to everything within it (multiplying any existing m-factors on those lower level blocks by the m-factor on the instance).

You can add:

outPar info what=output where=file

into the netlist, and then you can see the m-factor on each of the lower level effective parameters (it will write a file called netlistSuffix.info.output into the working dir with the information for the output parameters of each device).

Andrew.

Sorry, forgot to say that this can be done even if the instance does not already have m on it.

Andrew.

Thanks Andrew, 

I used this command line  :trim1 alter dev=X8SEG\<0\>.XSEG\<7\> param=m value=4

and I have this ERROR:

`trim1': Sweeping the parameters of a subcircuit instance (`XSEG<7>') is currently not supported by `dev' parameter. Please use `sub' parameter instead.

So, I replaced 'dev' by 'sub', as said in the ERROR and now it works with no problem.

Thanks again, I am really grateful for your continuous support and smart solutions.

Regards,

kotb

Hi Andrew,

When I use   trim1 alter sub=X8SEG\<0\>.XSEG\<7\> param=m value=4  , I have no warning in the log file and the command line is taken in consideration because when I look into netlistSuffix.info.output, I find that all devices in the instance X8SEG\<0\>.XSEG\<7\>  have a multiplicity value meff=4

But when I see simulation results, nothing changes even I put param=m value=160 , just to have a huge load. the value of m is always equal to the initial value in the netlist which is m=1.

I find your answer in some problem which looks a little bit similar in this link

https://community.cadence.com/cadence_technology_forums/f/custom-ic-design/23118/spectre-parameters-in-alter-statement-not-effecting-digital-vector-file-stimuli

Do you think it is the same problem of spectre control mode (interactive or batch)? If so, could you please tell me what should I change to tell spectre that I am using batch mode ?

Regards,

kotb

If this is an issue, please contact customer support. I don't think there's anything in the post you referenced which is relevant here, and the workaround of using controlMode batch isn't going to help - if this happens in the first simulation you run, it's not going to matter - plus Ultrasim is not run in interactive mode.

Andrew.