AMD/Spansion FLASH problem

N

Niv

Guest
I've downloaded the am29dl322d (and others in the am29 series) and
tried to simulate with them.
There seems to be a lot of problems with this (these) model(s).
Initially, the model wouldn't compile due to some incompatibility with
the ieee.vital_timing library, requiring 2 lines to be commented out in
the two output path generate statements at the end of the model. It
then compiled OK (thanks to Mentor). However, when I try to simulate,
I get an error immediately saying:

# ** Fatal: (vsim-3421) Value 11 is out of range 1 to 4.
# Time: 0 fs Iteration: 0 Process:
/fmc_tb/u_0/u_3/behavior/functional File:
J:/POOL/ELEC_DES/Meteor/FPGA_DESIGN/LOFR/VERIFICATION/SIMULATION/MODELS/MDL_05_FLASH/RTL_HDL/am29dl324d.vhd
# Fatal error at
J:/POOL/ELEC_DES/Meteor/FPGA_DESIGN/LOFR/VERIFICATION/SIMULATION/MODELS/MDL_05_FLASH/RTL_HDL/am29dl324d.vhd
line 2683

So, are these models really that bad or am I doing something wrong?
This is now holding uo my progress in designing an FPGA interface to
these FLASH device(s).

Regards, Kev P.
 
Niv wrote:

I've downloaded the am29dl322d (and others in the am29 series) and
tried to simulate with them.
Click to expand...
Downloaded from where?

--
Paul.
www.aimcom.nl
email address: switch x and s
 
Kev,

The problem is that ModelSim by default points to an obsolete version
(VITAL'95) of the VITAL packages. Change your modelsim.ini file to:

ieee = $MODEL_TECH/../vital2000

It should compile without errors.

Rick Munden


Niv wrote:
I've downloaded the am29dl322d (and others in the am29 series) and
tried to simulate with them.
There seems to be a lot of problems with this (these) model(s).
Initially, the model wouldn't compile due to some incompatibility with
the ieee.vital_timing library, requiring 2 lines to be commented out in
the two output path generate statements at the end of the model. It
then compiled OK (thanks to Mentor). However, when I try to simulate,
I get an error immediately saying:

# ** Fatal: (vsim-3421) Value 11 is out of range 1 to 4.
# Time: 0 fs Iteration: 0 Process:
/fmc_tb/u_0/u_3/behavior/functional File:
J:/POOL/ELEC_DES/Meteor/FPGA_DESIGN/LOFR/VERIFICATION/SIMULATION/MODELS/MDL_05_FLASH/RTL_HDL/am29dl324d.vhd
# Fatal error at
J:/POOL/ELEC_DES/Meteor/FPGA_DESIGN/LOFR/VERIFICATION/SIMULATION/MODELS/MDL_05_FLASH/RTL_HDL/am29dl324d.vhd
line 2683

So, are these models really that bad or am I doing something wrong?
This is now holding uo my progress in designing an FPGA interface to
these FLASH device(s).

Regards, Kev P.
Click to expand...
 
Paul,

The models are available from the Spansion website and from
www.FreeModelFoundry.com



Paul Uiterlinden wrote:
Niv wrote:

I've downloaded the am29dl322d (and others in the am29 series) and
tried to simulate with them.

Downloaded from where?
Click to expand...
 
FMF wrote:
Kev,

The problem is that ModelSim by default points to an obsolete version
(VITAL'95) of the VITAL packages. Change your modelsim.ini file to:

ieee = $MODEL_TECH/../vital2000

It should compile without errors.

Rick Munden


Niv wrote:
I've downloaded the am29dl322d (and others in the am29 series) and
tried to simulate with them.
There seems to be a lot of problems with this (these) model(s).
Initially, the model wouldn't compile due to some incompatibility with
the ieee.vital_timing library, requiring 2 lines to be commented out in
the two output path generate statements at the end of the model. It
then compiled OK (thanks to Mentor). However, when I try to simulate,
I get an error immediately saying:

# ** Fatal: (vsim-3421) Value 11 is out of range 1 to 4.
# Time: 0 fs Iteration: 0 Process:
/fmc_tb/u_0/u_3/behavior/functional File:
J:/POOL/ELEC_DES/Meteor/FPGA_DESIGN/LOFR/VERIFICATION/SIMULATION/MODELS/MDL_05_FLASH/RTL_HDL/am29dl324d.vhd
# Fatal error at
J:/POOL/ELEC_DES/Meteor/FPGA_DESIGN/LOFR/VERIFICATION/SIMULATION/MODELS/MDL_05_FLASH/RTL_HDL/am29dl324d.vhd
line 2683

So, are these models really that bad or am I doing something wrong?
This is now holding uo my progress in designing an FPGA interface to
these FLASH device(s).

Regards, Kev P.
Click to expand...
It would seem I've finally got the VHDl FLASH model to work!
I was editing the entity generics rather than assigning them to the
files they needed where the model was instantiated, but somehow my
editing of the source did'nt come through to the instantiated file
(elbeit 2 of the 3 files are empty).

Just need to see if my sim actually works, now that the model doesn't
crash ModelSim.

A bit of a learning curve finally climbed!

Kev P.
 
The "full" statement spans five lines:

IF TimingModel(11) = 't' THEN
CFI_array(16#4F#) := 16#03#;
ELSE
CFI_array(16#4F#) := 16#02#;
END IF;

I can't tell what value the generic "TimingModel" was set to but
apparently it was a string only 4 characters.


Paul Uiterlinden wrote:
FMF wrote:

Paul,

The models are available from the Spansion website and from
www.FreeModelFoundry.com

I guessed as much. But then the line number of the error as reported
by the OP does not make much sense to me:

line 2683: CFI_array(16#46#) := 16#02#;

How can this give a "Value 11 is out of range 1 to 4" error? That's
why I wondered whether I was looking at the correct file. Perhaps
this is one of the wonders of ModelSim's error reporting.
Click to expand...
 
FMF wrote:

The "full" statement spans five lines:

IF TimingModel(11) = 't' THEN
CFI_array(16#4F#) := 16#03#;
ELSE
CFI_array(16#4F#) := 16#02#;
END IF;

I can't tell what value the generic "TimingModel" was set to but
apparently it was a string only 4 characters.
Click to expand...
But that is line 2692. The OP reported line 2683. Are we talking about
the same file here?

-- File name : am29dl324d.vhd
--
-- version: | author: | mod date: | changes made:
-- V1.0 J.Bogosavljevic 04 Apr 28 - initial release

That is the one I donwloaded from the web site.

--
Paul.
www.aimcom.nl
email address: switch x and s
 
Paul,

We are talking about the same file and you are right about the line
numbers. I guess I can't explain why ModelSim comes up with that line
number but I still suspect the actual problem is at line 2692.

It could be the original poster got his model from the Spansion website
which would have a shorter copyright statement at the top. Line 2983 in
his file could be line 2692 in FMF's.


Paul Uiterlinden wrote:
FMF wrote:

The "full" statement spans five lines:

IF TimingModel(11) = 't' THEN
CFI_array(16#4F#) := 16#03#;
ELSE
CFI_array(16#4F#) := 16#02#;
END IF;

I can't tell what value the generic "TimingModel" was set to but
apparently it was a string only 4 characters.

But that is line 2692. The OP reported line 2683. Are we talking about
the same file here?

-- File name : am29dl324d.vhd
--
-- version: | author: | mod date: | changes made:
-- V1.0 J.Bogosavljevic 04 Apr 28 - initial release

That is the one I donwloaded from the web site.
Click to expand...
 
FMF wrote:
Paul,

We are talking about the same file and you are right about the line
numbers. I guess I can't explain why ModelSim comes up with that line
number but I still suspect the actual problem is at line 2692.

It could be the original poster got his model from the Spansion website
which would have a shorter copyright statement at the top. Line 2983 in
his file could be line 2692 in FMF's.



Yes, I downloaded it from the Spansion site, not from FMF.
Click to expand...

Pleased to report all is working fine now. Next battle will be to
understand how to get the .sdf file implemented & working.
 
FMF wrote:
Kev,

The problem is that ModelSim by default points to an obsolete version
(VITAL'95) of the VITAL packages. Change your modelsim.ini file to:

ieee = $MODEL_TECH/../vital2000

It should compile without errors.

Rick Munden


I tried pointing to the vital2000 lib, but they still failed to compile.
Click to expand...
 
FMF wrote:

Paul,

The models are available from the Spansion website and from
www.FreeModelFoundry.com
Click to expand...
I guessed as much. But then the line number of the error as reported
by the OP does not make much sense to me:

line 2683: CFI_array(16#46#) := 16#02#;

How can this give a "Value 11 is out of range 1 to 4" error? That's
why I wondered whether I was looking at the correct file. Perhaps
this is one of the wonders of ModelSim's error reporting.

--
Paul.
www.aimcom.nl
email address: switch x and s
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top