[SI-LIST] AW: Re: Run Scheduled Drivers IBIS models on HSPICE 2001.4


Hi Michael,

After I changed the IBIS files as I mentioned below=20
the simulation runs without any errors.=20

I didn't get the error message which occurred in your=20
simulation and I don't know how to handle it.=20
So, I'm sorry that I can't help you in this case.

For your information: I've performed the same simulation
with SpecctraQuest without modifying the IBIS file=20
and this runs without any errors.=20

Furthermore I've created an HSPICE bug enhancement on=20
Jan-29-2002 which is not yet closed.=20

Sorry, that I can't help you more.
Dirk.


> -----Urspr=FCngliche Nachricht-----
> Von: Chan, Michael [mailto:Michael.Chan@xxxxxxxxxx]
> Gesendet: Freitag, 19. April 2002 00:02
> An: Linnenbruegger Dirk
> Betreff: RE: [SI-LIST] Re: Run Scheduled Drivers IBIS models on =
HSPICE
> 2001.4
>=20
>=20
> Dirk:
>      I follow your advices and change the two scheduled=20
> drivers to I/O_open_drain and I/O_open_source.
> The error message of "inconsistent nodelist ###### " now=20
> disappears but it generates a new error
> " parsing a parameter and could not find the =3D  "  message.=20
> Have you see similar error at your side.=20
> I understand you have no obligation to look at this further=20
> but I really appreciate all your help.
> I do check all the details inside the *.ibs file and I cannot=20
> see any error inside that could trigger
> this error.=20
>=20
> BTW I run the same *.ibs file ( without adding I/O to the=20
> schedule driver model type ) under HYPERLYNX
> and it runs without problem at all.
>=20
> Regards,
> Michael
>=20
> -----Original Message-----
> From: Linnenbruegger Dirk
> [mailto:Dirk.Linnenbruegger@xxxxxxxxxxxxxxxxxxx]
> Sent: Thursday, April 18, 2002 2:48 AM
> To: 'tracyb@xxxxxxxxxxxxxx'; Chan, Michael
> Cc: si-list@xxxxxxxxxxxxx
> Subject: [SI-LIST] Re: Run Scheduled Drivers IBIS models on HSPICE
> 2001.4
>=20
>=20
>=20
> Hi Michael,
>=20
> I've had the same problems with the [DRIVER_SCHEDULE]
> devices as you mentioned.
>=20
>=20
> After I had a closer look to the IBIS file I found out=20
> that the top level model is an I/O open drain buffer and=20
> the scheduled drivers are open drain (MODEL1_N ) and=20
> open source buffers (MODEL1_P).
>=20
> When I use the following buffer syntax (all buffers use
> the same nodelist -as Tracy mentioned-), Hspice always=20
> aborts with:
>=20
> ******  Star-HSPICE -- 2001.4   **************************
> b_BUFFER_io  BUFFER_pu  BUFFER_pd  =20
>    +         BUFFER_out BUFFER_in    =20
>    +         BUFFER_nd_en  BUFFER_V_out_of_in =20
> + file     =3D '../Lib/model1.ibs'
> + model    =3D 'MODEL1'
> + buffer   =3D 6     $ I/O open drain (top level model)
> **error** Inconsisitent nodelist between the top and the=20
> scheduled model where File =3D ../Lib/model1.ibs Model =3D MODEL1
>=20
>=20
>=20
> My workaround to solve this problem is to change the buffer=20
> type of the scheduled buffers in the IBIS file in the=20
> following way:
>=20
> [Model]     MODEL1_N =20
> Model_type  open_drain -> I/O_open_drain
>=20
> [Model]     CPU_MID_MODEL1_P
> Model_type  open_source -> I/O_open_source
>=20
>=20
> Maybe this will help you.
>=20
> Dirk Linnenbruegger
> Fujitsu Siemens Computers GmbH
>=20
>=20
> > -----Urspr=FCngliche Nachricht-----
> > Von: Tracy Barclay [mailto:tracyb@xxxxxxxxxxxxxx]
> > Gesendet: Mittwoch, 17. April 2002 20:34
> > An: Michael.Chan@xxxxxxxxxx
> > Cc: si-list@xxxxxxxxxxxxx
> > Betreff: [SI-LIST] Re: Run Scheduled Drivers IBIS models on HSPICE
> > 2001.4
> >=20
> >=20
> >=20
> > Hi Michael,
> >=20
> > You do not have to add anything into your netlist in order=20
> for [Driver
> > Schedule] to work.  Hspice reads the [Driver Schedule] data and the
> > corresponding 'scheduled' models from the IBIS file.  In the=20
> > IBIS file, the
> > [Driver Schedule] data will look something like:
> >=20
> > [Driver Schedule]
> > | Model_name  Rise_on_dly  Rise_off_dly  Fall_on_dly  Fall_off_dly
> > CPU_N                  0                    NA
> > 0                 NA
> > CPU_P               0.250ns           2.5ns           NA     =20
> >           NA
> >=20
> > In this example, the scheduled models are CPU_N and CPU_P. =20
> > The models must
> > be of the same model type as the top level model that these=20
> > are scheduled
> > for  So, if the top level model is model type I/O, then both=20
> > the scheduled
> > models CPU_N and CPU_P must also be I/O buffers.  If the=20
> > scheduled buffers
> > are not of the same model type as the top level buffer, you=20
> > will get a error
> > message, "Inconsistent nodelist between the top and the=20
> > scheduled model" and
> > Hspice will abort.
> >=20
> > The driver schedule implementation will be improved in the=20
> > next release so
> > that the scheduled models do not have to be the same model=20
> > type as the top
> > level model.
> >=20
> > Let me know if you have any other questions about this.
> >=20
> > Tracy Barclay
> > Star-Hspice Product Specialist
> > Avant! Corporation
> >=20
> > > Gurus:
> > >           I have a scheduled drivers IBIS model that I try=20
> > to run on =3D
> > > HSPICE 2001.4. HSPICE manual does not seems to have a=20
> > nodelist format =3D
> > > for=3D20
> > > the scheduled drivesr IBIS model. Am I supposed to use the=20
> > nodelist as a =3D
> > > generic I/O buffer as described inside the manual or I have=20
> > to use a=3D20
> > > specific nodelist just for scheduled drivers ? I sent a=20
> request to =3D
> > > AVANTI but no response so far. Thanks.
> > >
> > > Regards,
> > > Michael Chan
> > >
> > > =
------------------------------------------------------------------
>=20
------------------------------------------------------------------
To unsubscribe from si-list:
si-list-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject field

or to administer your membership from a web page, go to:
http://www.freelists.org/webpage/si-list

For help:
si-list-request@xxxxxxxxxxxxx with 'help' in the Subject field

List archives are viewable at:     
                http://www.freelists.org/archives/si-list
or at our remote archives:
                http://groups.yahoo.com/group/si-list/messages 
Old (prior to June 6, 2001) list archives are viewable at:
                http://www.qsl.net/wb6tpu

Other related posts: