This problem was solved. The problem
was caused by use of old CASPER
libraries (version 10.1) which lack
good support for David's procedure.
Upgrading to a recent version (and
also upgrading Xilinx DSP stuff to
match) solved the problem.

> The pcore block is in the model, the
> names seem to be the same from the
> original blackboxpfb2.mdl file to its
> use in the testblackbox.mdl file, and
> the names and widths in the VHDL file
> seem to match the original mdl file.
> I tried using relative paths to the
> NGC and VHDL files and then absolute
> paths without differences in results.
> The reason I attached the zipped tar
> file to my previous email is I think
> there may be some other subtle error.
> Did you see any problem?
>
>> I thought I stayed inside the rules.
>> Attached is a zipped tar file with the
>> blackboxpfg2.mdl and testblackbox.mdl
>> files of my second attempt. Also in the
>> tar file are the VHDL and Matlab files
>> generated by the procedure. Do you see
>> any violations?
>>
>>> A few possibilities come to mind:
>>>
>>> 1) Did you remember to add a "pcore" block to the model (see section
>>> 3.4
>>> of the Black Boxing article)?
>>>
>>> 2) Have you changed the name or width of a Gateware In or Gateway Out
>>> block in the model that defines the netlist?  If so, you might need to
>>> have the block box wizard regenerate the *_core_config.m file for you
>>> (or
>>> make the appropriate edits manually if you prefer).
>>>
>>> 3) Make sure that the port names and widths listed in the *_core.vhd
>>> file
>>> (in your top level directory), and in the *_core_config.m file, and in
>>> the
>>> *_core.mdl file all match.  Watch out for Gateway In/Out blocks whose
>>> name
>>> is a VHDL reserved word.
>>>
>>> Hope this helps,
>>> Dave
>>>
>>> On Jan 30, 2012, at 5:52 PM, [email protected] wrote:
>>>
>>>> I am having trouble with my Black Boxes.
>>>> Compile times for my models are so long
>>>> that I need to use the Xilinx Black Box
>>>> feature. I have the article by "Optimize
>>>> CASPER Development by 'Black Boxing'
>>>> Designs" by David MacMahon November 23,
>>>> 2010. David's procedure seems to easily
>>>> build black boxes, but when I use bee_xps
>>>> to compile simple models that use them, I
>>>> get the following dreaded messages.
>>>>
>>>>   ERROR:NgdBuild:604 - logical block
>>>>   'testblackbox_XSG_core_config/
>>>>   testblackbox_XSG_core_config/testblackbox_x0/
>>>>   my_blackboxpfb_core' with type 'blackboxpfb_core'
>>>>   could not be resolved. A pin name misspelling can
>>>>   cause this, a missing edif or ngc file, or the
>>>>   misspelling of a type name. Symbol
>>>>   'blackboxpfb_core' is not supported in target
>>>>   'virtex5'.
>>>>
>>>> Making my "blackboxpfb" black box
>>>> created a subdirectory "blackboxpfb_core"
>>>> with the following contents.
>>>>
>>>> addsb_11_0_bdb276565dee6659.ngc  bmg_33_9d3bf13e9a4d9a01.mif
>>>> blackboxpfb_core_cw.gise         bmg_33_9d3bf13e9a4d9a01.ngc
>>>> blackboxpfb_core_cw_import.log   bmg_33_9fed4de7958f9a58.mif
>>>> blackboxpfb_core_cw_import.tcl   bmg_33_9fed4de7958f9a58.ngc
>>>> blackboxpfb_core_cw.ise          bmg_33_a33051899a02b0df.mif
>>>> blackboxpfb_core_cw.sdc          bmg_33_a33051899a02b0df.ngc
>>>> blackboxpfb_core_cw.sgp          bmg_33_a38853f09a992471.mif
>>>> blackboxpfb_core_cw.ucf          bmg_33_a38853f09a992471.ngc
>>>> blackboxpfb_core_cw.vhd          bmg_33_a472e9469e1a55bc.mif
>>>> blackboxpfb_core_cw.xcf          bmg_33_a472e9469e1a55bc.ngc
>>>> blackboxpfb_core_cw_xdb          bmg_33_b433f7e31442d1b4.mif
>>>> blackboxpfb_core_cw.xise         bmg_33_b433f7e31442d1b4.ngc
>>>> blackboxpfb_core.ngc             bmg_33_bde0d98b7c268d95.mif
>>>> blackboxpfb_core.vhd             bmg_33_bde0d98b7c268d95.ngc
>>>> blackboxpfb_core.vho             bmg_33_be5a393c8320608f.mif
>>>> bmg_33_01dd252156a2c7b3.mif      bmg_33_be5a393c8320608f.ngc
>>>> bmg_33_01dd252156a2c7b3.ngc      bmg_33_c0ac3413361ea52b.mif
>>>> bmg_33_047c125a73f3c1f0.mif      bmg_33_c0ac3413361ea52b.ngc
>>>> bmg_33_047c125a73f3c1f0.ngc      bmg_33_c5560a636d03fc1c.mif
>>>> bmg_33_482b8d0961f68d52.mif      bmg_33_c5560a636d03fc1c.ngc
>>>> bmg_33_482b8d0961f68d52.ngc      bmg_33_d4b4989bce18e25c.mif
>>>> bmg_33_4ed11faae7e5e6b4.mif      bmg_33_d4b4989bce18e25c.ngc
>>>> bmg_33_4ed11faae7e5e6b4.ngc      bmg_33_dff652364d3ec538.mif
>>>> bmg_33_55833db8ed9147a1.mif      bmg_33_dff652364d3ec538.ngc
>>>> bmg_33_55833db8ed9147a1.ngc      bmg_33_e1a35aaacd7ca869.mif
>>>> bmg_33_561be7213f0e7611.mif      bmg_33_e1a35aaacd7ca869.ngc
>>>> bmg_33_561be7213f0e7611.ngc      bmg_33_e5a5cae348665945.mif
>>>> bmg_33_5fb1618bdd9e2378.mif      bmg_33_e5a5cae348665945.ngc
>>>> bmg_33_5fb1618bdd9e2378.ngc      bmg_33_e6871bb194f4a37b.mif
>>>> bmg_33_6101afd66f01b855.mif      bmg_33_e6871bb194f4a37b.ngc
>>>> bmg_33_6101afd66f01b855.ngc      bmg_33_f375f5a3a02b6146.mif
>>>> bmg_33_7592cf5466cc0a10.mif      bmg_33_f375f5a3a02b6146.ngc
>>>> bmg_33_7592cf5466cc0a10.ngc      cntr_11_0_64b6c809edee1e81.ngc
>>>> bmg_33_7a66187dde5fb2db.mif      cntr_11_0_9cd1168730d276e8.ngc
>>>> bmg_33_7a66187dde5fb2db.ngc      commandLines
>>>> bmg_33_7b6922b2a0594c9d.mif      globals
>>>> bmg_33_7b6922b2a0594c9d.ngc      hdlFiles
>>>> bmg_33_8008c267f268c930.mif      isim_blackboxpfb_core.prj
>>>> bmg_33_8008c267f268c930.ngc      name_translations
>>>> bmg_33_8158159440be6c26.mif      SgIseProject.tcl
>>>> bmg_33_8158159440be6c26.ngc      synopsis
>>>> bmg_33_845cfe1c79250673.mif      synth_model
>>>> bmg_33_845cfe1c79250673.ngc      sysgen
>>>> bmg_33_8478501b21bb5c0f.mif      vcom.do
>>>> bmg_33_8478501b21bb5c0f.ngc      xlpersistentdff.ngc
>>>> bmg_33_8a3f1db33d02b6d7.mif      _xmsgs
>>>> bmg_33_8a3f1db33d02b6d7.ngc      xst_blackboxpfb_core.prj
>>>> bmg_33_8e043b8dc69da1e8.mif      xst_blackboxpfb_core.scr
>>>> bmg_33_8e043b8dc69da1e8.ngc
>>>>
>>>> My Matlab list of paths includes my
>>>> work directory and this subdirectory.
>>>> In my work directory, there is also a
>>>> "blackboxpfb_core.vhd" that was
>>>> generated by "extract_entity" in one
>>>> of the steps for making the black box.
>>>> Thus, there are two of these VHDL files,
>>>> one in my work directory and one in the
>>>> "blackboxpfb_core" subdirectory. The
>>>> two VHDL files differ, but were both
>>>> generated by the procedure given in
>>>> David's article. I have CASPER library
>>>> mlib_devel_10_1 and Xilinx 11.1
>>>> software. My Matlab version is 2009a.
>>>>
>>>> What did I do wrong?
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>>
>>
>
>
>



Reply via email to