CDL netlister... works in a wrong way!

C

camelot

Guest
Hello,
I have a question that's driving me crazy. A colleague sent me a
schematic (opus4.4.6) having a personalized symbol he created. Well,
when I netlist a view containing his custom symbol using netlister like
eldo or spectre all work fine. When I try to do a CDLout, the
netlisting is wrong. There are net connected in wrong pin name. In
order to investigate the problem, I created a standard new symbol for
that schematic and with the my symbol the netlisting is correct both
with eldo/spectre and cdl.
My doubt is that the devil symbol contains something strange that make
the cdl netlister crazy. Could someone indicate me something that
could help me?

Thank you,

Camelot
 
Have you checked the simulation information (siminfo) for cdl (auCdl) in the
Component Description Format (CDF) for the cell which holds the symbol.
Maybe something is messed up there, maybe 'termOrder'.

If you create a new symbol CDF will also created new.

Bernd


camelot wrote:
Hello,
I have a question that's driving me crazy. A colleague sent me a
schematic (opus4.4.6) having a personalized symbol he created. Well,
when I netlist a view containing his custom symbol using netlister like
eldo or spectre all work fine. When I try to do a CDLout, the
netlisting is wrong. There are net connected in wrong pin name. In
order to investigate the problem, I created a standard new symbol for
that schematic and with the my symbol the netlisting is correct both
with eldo/spectre and cdl.
My doubt is that the devil symbol contains something strange that make
the cdl netlister crazy. Could someone indicate me something that
could help me?

Thank you,

Camelot
 
Hi Bernd,
yes, I already check the simInfo parameters and it seems to be fine.
The cell contains inside only 6 resistors (the same) not connected
between them but only connrected with the 12 pins then presents in the
symbols. It's a simple block of resistors.

This is the simInfo information of the resistor cell:

netlistprocedure: ansCdlCompPrim
instParameter: L W R
componentName: res
termOrder: PLUS MINUS
propMapping: nil L length W width R r
name Prefix: R
modelName NR

Could I check anything else?

Camelot

Bernd Fischer ha scritto:

Have you checked the simulation information (siminfo) for cdl (auCdl) in the
Component Description Format (CDF) for the cell which holds the symbol.
Maybe something is messed up there, maybe 'termOrder'.

If you create a new symbol CDF will also created new.

Bernd


camelot wrote:
Hello,
I have a question that's driving me crazy. A colleague sent me a
schematic (opus4.4.6) having a personalized symbol he created. Well,
when I netlist a view containing his custom symbol using netlister like
eldo or spectre all work fine. When I try to do a CDLout, the
netlisting is wrong. There are net connected in wrong pin name. In
order to investigate the problem, I created a standard new symbol for
that schematic and with the my symbol the netlisting is correct both
with eldo/spectre and cdl.
My doubt is that the devil symbol contains something strange that make
the cdl netlister crazy. Could someone indicate me something that
could help me?

Thank you,

Camelot
 
ops... sorry..I forgot to told that the top cell that holds the
resistors does not has any CDF information. The auCdl simInfo is blank.
It should be correct, no?

Camelot


camelot ha scritto:

Hi Bernd,
yes, I already check the simInfo parameters and it seems to be fine.
The cell contains inside only 6 resistors (the same) not connected
between them but only connrected with the 12 pins then presents in the
symbols. It's a simple block of resistors.

This is the simInfo information of the resistor cell:

netlistprocedure: ansCdlCompPrim
instParameter: L W R
componentName: res
termOrder: PLUS MINUS
propMapping: nil L length W width R r
name Prefix: R
modelName NR

Could I check anything else?

Camelot

Bernd Fischer ha scritto:

Have you checked the simulation information (siminfo) for cdl (auCdl) in the
Component Description Format (CDF) for the cell which holds the symbol.
Maybe something is messed up there, maybe 'termOrder'.

If you create a new symbol CDF will also created new.

Bernd


camelot wrote:
Hello,
I have a question that's driving me crazy. A colleague sent me a
schematic (opus4.4.6) having a personalized symbol he created. Well,
when I netlist a view containing his custom symbol using netlister like
eldo or spectre all work fine. When I try to do a CDLout, the
netlisting is wrong. There are net connected in wrong pin name. In
order to investigate the problem, I created a standard new symbol for
that schematic and with the my symbol the netlisting is correct both
with eldo/spectre and cdl.
My doubt is that the devil symbol contains something strange that make
the cdl netlister crazy. Could someone indicate me something that
could help me?

Thank you,

Camelot
 
I solved the problem. In my .simrc I found this variable set to true:
auCdlCDFPinCntrl = t
I deleted this setting and all worked fine.

....but does anybody know what this variable do exactly?

Camelot


camelot ha scritto:

ops... sorry..I forgot to told that the top cell that holds the
resistors does not has any CDF information. The auCdl simInfo is blank.
It should be correct, no?

Camelot


camelot ha scritto:

Hi Bernd,
yes, I already check the simInfo parameters and it seems to be fine.
The cell contains inside only 6 resistors (the same) not connected
between them but only connrected with the 12 pins then presents in the
symbols. It's a simple block of resistors.

This is the simInfo information of the resistor cell:

netlistprocedure: ansCdlCompPrim
instParameter: L W R
componentName: res
termOrder: PLUS MINUS
propMapping: nil L length W width R r
name Prefix: R
modelName NR

Could I check anything else?

Camelot

Bernd Fischer ha scritto:

Have you checked the simulation information (siminfo) for cdl (auCdl) in the
Component Description Format (CDF) for the cell which holds the symbol.
Maybe something is messed up there, maybe 'termOrder'.

If you create a new symbol CDF will also created new.

Bernd


camelot wrote:
Hello,
I have a question that's driving me crazy. A colleague sent me a
schematic (opus4.4.6) having a personalized symbol he created. Well,
when I netlist a view containing his custom symbol using netlister like
eldo or spectre all work fine. When I try to do a CDLout, the
netlisting is wrong. There are net connected in wrong pin name. In
order to investigate the problem, I created a standard new symbol for
that schematic and with the my symbol the netlisting is correct both
with eldo/spectre and cdl.
My doubt is that the devil symbol contains something strange that make
the cdl netlister crazy. Could someone indicate me something that
could help me?

Thank you,

Camelot
 
cdsdoc says

auCdlCDFPinCntrlGo = 't
Allows CDF termOrder to dictate pin ordering of the top-level cell or the cell
that has the auCdl view. The default is 'nil.

Bernd


camelot wrote:
I solved the problem. In my .simrc I found this variable set to true:
auCdlCDFPinCntrl = t
I deleted this setting and all worked fine.

...but does anybody know what this variable do exactly?

Camelot
 
But if the cell, as in my case, does not have CDF simulation
information what happens?

Camelot


Bernd Fischer ha scritto:

cdsdoc says

auCdlCDFPinCntrlGo = 't
Allows CDF termOrder to dictate pin ordering of the top-level cell or the cell
that has the auCdl view. The default is 'nil.

Bernd


camelot wrote:
I solved the problem. In my .simrc I found this variable set to true:
auCdlCDFPinCntrl = t
I deleted this setting and all worked fine.

...but does anybody know what this variable do exactly?

Camelot
 
On 22 Sep 2006 23:18:18 -0700, "camelot" <scosmo@tiscalinet.it> wrote:

But if the cell, as in my case, does not have CDF simulation
information what happens?

Camelot


Bernd Fischer ha scritto:

cdsdoc says

auCdlCDFPinCntrlGo = 't
Allows CDF termOrder to dictate pin ordering of the top-level cell or the cell
that has the auCdl view. The default is 'nil.

Bernd


camelot wrote:
I solved the problem. In my .simrc I found this variable set to true:
auCdlCDFPinCntrl = t
I deleted this setting and all worked fine.

...but does anybody know what this variable do exactly?

Camelot
Nowadays it gives a warning, and uses the default terminal order (if my memory
is correct), It checks that the termOrder is present in the CDF, and consistent
with the terminals of the top level cellView.

Regards,

Andrew.
--
Andrew Beckett
Principal European Technology Leader
Cadence Design Systems, UK.
 
Open up the symbol, edit-->properties-->port order.

this will pop a panel and you can change the netlisting order of the
block.

Rick


Andrew Beckett wrote:
On 22 Sep 2006 23:18:18 -0700, "camelot" <scosmo@tiscalinet.it> wrote:

But if the cell, as in my case, does not have CDF simulation
information what happens?

Camelot


Bernd Fischer ha scritto:

cdsdoc says

auCdlCDFPinCntrlGo = 't
Allows CDF termOrder to dictate pin ordering of the top-level cell or the cell
that has the auCdl view. The default is 'nil.

Bernd


camelot wrote:
I solved the problem. In my .simrc I found this variable set to true:
auCdlCDFPinCntrl = t
I deleted this setting and all worked fine.

...but does anybody know what this variable do exactly?

Camelot


Nowadays it gives a warning, and uses the default terminal order (if my memory
is correct), It checks that the termOrder is present in the CDF, and consistent
with the terminals of the top level cellView.

Regards,

Andrew.
--
Andrew Beckett
Principal European Technology Leader
Cadence Design Systems, UK.
 
Open up the symbol, edit-->properties-->port order.

this will pop a panel and you can change the netlisting order of the
block.

Rick


Andrew Beckett wrote:
On 22 Sep 2006 23:18:18 -0700, "camelot" <scosmo@tiscalinet.it> wrote:

But if the cell, as in my case, does not have CDF simulation
information what happens?

Camelot


Bernd Fischer ha scritto:

cdsdoc says

auCdlCDFPinCntrlGo = 't
Allows CDF termOrder to dictate pin ordering of the top-level cell or the cell
that has the auCdl view. The default is 'nil.

Bernd


camelot wrote:
I solved the problem. In my .simrc I found this variable set to true:
auCdlCDFPinCntrl = t
I deleted this setting and all worked fine.

...but does anybody know what this variable do exactly?

Camelot


Nowadays it gives a warning, and uses the default terminal order (if my memory
is correct), It checks that the termOrder is present in the CDF, and consistent
with the terminals of the top level cellView.

Regards,

Andrew.
--
Andrew Beckett
Principal European Technology Leader
Cadence Design Systems, UK.
 

Welcome to EDABoard.com

Sponsor

Back
Top