Bug in DDR template in Lattice FPGAs ?

A

ALuPin

Hi,

I am trying to use the "DDR_MEM" Lattice template which is responsable
for the datapath for DDR SDRAM controller.

"DDR_MEM" can be found in the MODULE/IP MANAGER under
ARCHITECTURE_MODULES
--> IO --> DDR_MEM

When instantiating that module and compiling my design I can see in
the timing analysis
that the bits on the bidirectional busses DQ and DQS have
different Clock-To-Output times.

In my opinion the busses should be routed into IO register cells
when instantiating that special template.

So there are three possibilities:

1. They are not routed into IO register cells so that the PERFORMANCE
ANALYST does show different tCO
2. They ARE routed into IO register cells, but the PERFORMANCE ANALYST
does not take them into timing calculation
3. I have to make any assignment so that the busses are routed into
IO registers. But I have not found any assignment possibility in
the PREFERENCE EDITOR.

Has someone experienced similar or same problems ?

Thank you in advance.

Rgds
André
 
W

Weng Tianxiang

DQ and DQS should have different clock-to-output timing. DQS rising
edge should be in the middle of DQ output window. So receiver can use
DQS rising edge to clock in data contained in DQ valid window.

What DDR is different from normal SDRAM is DDR is source synchronous
driving device. DQS is used to clock in data sent by data sender.

Weng
 
A

ALuPin

Weng Tianxiang said:
DQ and DQS should have different clock-to-output timing. DQS rising
edge should be in the middle of DQ output window. So receiver can use
DQS rising edge to clock in data contained in DQ valid window.

What DDR is different from normal SDRAM is DDR is source synchronous
driving device. DQS is used to clock in data sent by data sender.

Weng

I am NOT talking of DQ with regard to DQS.

I am talking about the bits of the bus DQ and the bits of the bus
DQS ! They have different tCO.
I already know DDR basics ;o)
 
J

John_H

I am NOT talking of DQ with regard to DQS.

I am talking about the bits of the bus DQ and the bits of the bus
DQS ! They have different tCO.
I already know DDR basics ;o)

I'm still not answering your question but posing you another question:

Are the clock-to-out times different by about 1/4 of your DDR clock period?

I know that timing analysis from other vendors tend to illustrate timing
from the positive edge of the appropriate clock. If the Lattice
implementation uses a 2x clock for the data signals, the DQ "bus" and the
DQS "bus" (neither of which are a bus, really - they're coupled into byte
lanes) would be clocked on opposite edges of that clock. If the
clock-to-out timing is referenced to the rising edge only, the difference
would be half the 2x clock period. Is that what you're seeing?

Either a Lattice FAE or timing report numbers published here might give
others a better idea of what's happening.
 
A

ALuPin

OK, maybe my question was not clear enough, one more try:

Having a look at the data bus DQ I can see the following in the
PERFORMANCE ANALYST (that is the STATIC TIMING ANALYSIS):

SOURCE DESTINATION DELAY (ns)

CLK Dq_15 9.971
CLK Dq_14 9.971
CLK Dq_13 9.701
CLK Dq_12 9.701
CLK Dq_11 9.698
CLK Dq_10 9.698
CLK Dq_9 9.974
CLK Dq_8 9.974
CLK Dq_7 10.574
CLK Dq_6 10.070
CLK Dq_5 10.350
CLK Dq_4 10.314
CLK Dq_3 10.070
CLK Dq_2 10.574
CLK Dq_1 10.350
CLK Dq_0 10.314


As you can see there are different tCO for the bits of the bus,
the same for DQS, ALTHOUGH using the Lattice datapath template as
described in my previous post.

Any ideas?

Rgds
André
 
C

cas7406

Andre,

I think what you are missing are the constraints. Have set your Tco,
fmax, Setup and Pin Assignment constraints? Use either the preference
file (.prf) or the Pre-Map Preference Editor to set your constraints.
Also TN1050 can help you out.

rgds,

cristian
 
A

ALuPin

Andre,

I think what you are missing are the constraints. Have set your Tco,
fmax, Setup and Pin Assignment constraints? Use either the preference
file (.prf) or the Pre-Map Preference Editor to set your constraints.
Also TN1050 can help you out.

rgds,

cristian

Hi Cristian,

yes I have set the constraints for all the other signals which do not
come out the DATAPATH template. In the Timing Analysis these signals
(for example the address bus for DDR) do have the same tCO.

I have constrained DQ and DQS in the PREFERENCE EDITOR under "In/Out Clock"
whereas the other signals are constrained under "Cell Attributes".
But the constraints for DQ and DQS do have no effect on the tCO.

Rgds
André
 
C

cas7406

Andre,

You should be able to constraint the Tco for the DQ and DQS in the
input_setup/clock_to_output preference window.
Check the example at
$ispLEVER/ispcpld/examples/latticeEC/preferences_attributes/ddr/vhdl/ddr.syn

and let me know.

rgds,

cristian
 
A

ALuPin

Andre,

You should be able to constraint the Tco for the DQ and DQS in the
input_setup/clock_to_output preference window.
Check the example at
$ispLEVER/ispcpld/examples/latticeEC/preferences_attributes/ddr/vhdl/ddr.syn

and let me know.

rgds,

cristian

Hi Cristian,

I checked the example, and yet, the tCO is for the bits of DQ
identical.

But the example does not use the template from the IP Manager but
the particular modules.

ispLEVER5.0 is my hope :eek:)

Rgds
André
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Members online

No members online now.

Forum statistics

Threads
473,743
Messages
2,569,478
Members
44,899
Latest member
RodneyMcAu

Latest Threads

Top