05-10-2010 12:27 PM
Still the map problem..........
Updating timing models...
ERROR:Pack:2310 - Too many comps of type "RAMB36_EXP" found to fit this device.
ERROR:Map:237 - The design is too large to fit the device. Please check the Design Summary section to see which resource requirement for your design exceeds the resources available in the device. Note that the number of slices reported may not be reflected accurately as their packing might not have been completed.
NOTE: An NCD file will still be generated to allow you to examine the mapped
design. This file is intended for evaluation use only, and will not process
successfully through PAR.
How to solve this problem?
05-10-2010 12:37 PM
I'm using FIFO Generator v5.1 in the design part.
As the datasheet's introduction, FIFO depths could up to 4,194,304.
Now I change the FIFO depth from 65536 to 262144. I don't know why these kind of errors output.........
05-10-2010 02:29 PM
The error message seemed pretty clear to me. The device you are using does not
have enough block RAM bits to implement the FIFOs you have generated. Even one
of these FIFOs would require 4,194,304 bits. That would mean 128 36K block RAM's.
Does your part have 128 36K block RAMs?
05-10-2010 02:55 PM
Sorry. I'm trying to modify the design made by other guys.
1. Any way to check the block RAM's number?
2. Any other way to solve the problem? Such as using the Distributed RAM?
05-10-2010 03:10 PM
If you use the ISE Navigator, you can find the map report via "Design Summary/Reports" which looks
like a green greek letter sigma in the tool bar. Otherwise open up your map report file, which will
have a .mrp suffix. This should list all of the resources used in the design as well as how many
are available in the device. In the case of your block RAMs it should indicate that they are
"overmapped" and you can see by how much. Generally speaking, unless you are using
a very large device with a lot of the fabric unused, you won't be able to make up the difference
in RAM bits using distributed memory. Also note that very large distributed memories will
use a lot of routing resources and run very slow.