Cortex-M0 DesignStart processor size (FPGA)?

Dear sirs,

I have a question about the size of synthesized logic for Cortex-M0 DesignStart processor.

When I use ISE of Xilinx to synthesize Cortex-M0 DesignStart processor (version r1p0-00rel0), I have result as below.

(I configure for using Spartan-3AN, XC3S200AN, Grade = -4 in ISE).

Cortex-M0_processor.png

As Xilinx's mention at this link http://www.xilinx.com/publications/archives/xcell/Xcell64.pdf , they used around 3,000 LUT4

for Cortex-M1 (1,900 LUT4 at page 59 + 1,100 LUT4 at page 56 if Cortex-M1 included MMU block).

Have you ever synthesized Cortex-M0 DesignStart processor by ISE of Xilinx? I want to compare my result with yours. Because I need

to map the synthesized gate into Spartan-3AN XC3S200AN, but the overmapped occurred at 4 input LUT4 and Slices as my

attached picture.

If you have any idea for this, please send me in this topic.

PS: I am not know my post is suitable in this area or not. If it is not suitable, you can move it to another one and inform me the new

link.

Thanks.

Parents
  • Hello Jack, sorry that no-one got back to you on this question in a timely fashion.

    Yes, the design is too large for that device. You would need to move up to a larger capacity part.

    Using the requested device:

    Design Summary:
    Number of errors: 2
    Number of warnings: 4
    Logic Utilization:
    Number of Slice Flip Flops: 845 out of 3,840 22%
    Number of 4 input LUTs: 4,888 out of 3,840 127% (OVERMAPPED)
    Logic Distribution:
    Number of occupied Slices: 2,495 out of 1,920 129% (OVERMAPPED)
    Number of Slices containing only related logic: 2,495 out of 2,495 100%
    Number of Slices containing unrelated logic: 0 out of 2,495 0%
    *See NOTES below for an explanation of the effects of unrelated logic.
    Total Number of 4 input LUTs: 4,989 out of 3,840 129% (OVERMAPPED)
    Number used as logic: 4,888
    Number used as a route-thru: 101

    The Slice Logic Distribution report is not meaningful if the design is
    over-mapped for a non-slice resource or if Placement fails.

    Number of bonded IOBs: 136 out of 141 96%
    Number of BUFGMUXs: 1 out of 8 12%

    Average Fanout of Non-Clock Nets: 3.77

    Using a larger device:

    Design Information
    ------------------
    Command Line : map -intstyle ise -p xc3s400-pq208-4 -cm area -ir off -pr off
    -c 100 -o CORTEXM0DS_WRAPPER_map.ncd CORTEXM0DS_WRAPPER.ngd
    CORTEXM0DS_WRAPPER.pcf
    Target Device : xc3s400
    Target Package : pq208
    Target Speed : -4
    Mapper Version : spartan3 -- $Revision: 1.55 $
    Mapped Date : Wed Mar 15 17:04:42 2017

    Design Summary
    --------------
    Number of errors: 0
    Number of warnings: 5
    Logic Utilization:
    Number of Slice Flip Flops: 841 out of 7,168 11%
    Number of 4 input LUTs: 4,930 out of 7,168 68%
    Logic Distribution:
    Number of occupied Slices: 3,055 out of 3,584 85%
    Number of Slices containing only related logic: 3,055 out of 3,055 100%
    Number of Slices containing unrelated logic: 0 out of 3,055 0%
    *See NOTES below for an explanation of the effects of unrelated logic.
    Total Number of 4 input LUTs: 5,030 out of 7,168 70%
    Number used as logic: 4,930
    Number used as a route-thru: 100

    The Slice Logic Distribution report is not meaningful if the design is
    over-mapped for a non-slice resource or if Placement fails.

    Number of bonded IOBs: 136 out of 141 96%
    Number of BUFGMUXs: 1 out of 8 12%

    Average Fanout of Non-Clock Nets: 3.81

    The slight discrepancies in the numbers are because the second design ran to completion, whereas the first died quite early.
Reply
  • Hello Jack, sorry that no-one got back to you on this question in a timely fashion.

    Yes, the design is too large for that device. You would need to move up to a larger capacity part.

    Using the requested device:

    Design Summary:
    Number of errors: 2
    Number of warnings: 4
    Logic Utilization:
    Number of Slice Flip Flops: 845 out of 3,840 22%
    Number of 4 input LUTs: 4,888 out of 3,840 127% (OVERMAPPED)
    Logic Distribution:
    Number of occupied Slices: 2,495 out of 1,920 129% (OVERMAPPED)
    Number of Slices containing only related logic: 2,495 out of 2,495 100%
    Number of Slices containing unrelated logic: 0 out of 2,495 0%
    *See NOTES below for an explanation of the effects of unrelated logic.
    Total Number of 4 input LUTs: 4,989 out of 3,840 129% (OVERMAPPED)
    Number used as logic: 4,888
    Number used as a route-thru: 101

    The Slice Logic Distribution report is not meaningful if the design is
    over-mapped for a non-slice resource or if Placement fails.

    Number of bonded IOBs: 136 out of 141 96%
    Number of BUFGMUXs: 1 out of 8 12%

    Average Fanout of Non-Clock Nets: 3.77

    Using a larger device:

    Design Information
    ------------------
    Command Line : map -intstyle ise -p xc3s400-pq208-4 -cm area -ir off -pr off
    -c 100 -o CORTEXM0DS_WRAPPER_map.ncd CORTEXM0DS_WRAPPER.ngd
    CORTEXM0DS_WRAPPER.pcf
    Target Device : xc3s400
    Target Package : pq208
    Target Speed : -4
    Mapper Version : spartan3 -- $Revision: 1.55 $
    Mapped Date : Wed Mar 15 17:04:42 2017

    Design Summary
    --------------
    Number of errors: 0
    Number of warnings: 5
    Logic Utilization:
    Number of Slice Flip Flops: 841 out of 7,168 11%
    Number of 4 input LUTs: 4,930 out of 7,168 68%
    Logic Distribution:
    Number of occupied Slices: 3,055 out of 3,584 85%
    Number of Slices containing only related logic: 3,055 out of 3,055 100%
    Number of Slices containing unrelated logic: 0 out of 3,055 0%
    *See NOTES below for an explanation of the effects of unrelated logic.
    Total Number of 4 input LUTs: 5,030 out of 7,168 70%
    Number used as logic: 4,930
    Number used as a route-thru: 100

    The Slice Logic Distribution report is not meaningful if the design is
    over-mapped for a non-slice resource or if Placement fails.

    Number of bonded IOBs: 136 out of 141 96%
    Number of BUFGMUXs: 1 out of 8 12%

    Average Fanout of Non-Clock Nets: 3.81

    The slight discrepancies in the numbers are because the second design ran to completion, whereas the first died quite early.
Children
No data