3 replies to this topic

[xsound]

Grettings,

I created a FIR filter in CoDeveloper and exported it to Xilinx Generic VHDL (with floating point support). When I compiled the files, the program gave me an error saying that the floating point components were invalid. I poked around and found that the new version of the floating point libraries for Xilinx is 5.0. So I changed the version to 5. This is the library declaration in my code:

library XilinxCoreLib;
use XilinxCoreLib.floating_point_v5_0_comp.all;

and this is an example of the component created by CoDeveloper:

xilinx_fadd : floating_point_v5_0
generic map(
c_has_add => 1,
c_a_width => 32,
c_a_fraction_width => 24,
c_b_width => 32,
c_b_fraction_width => 24,
c_result_fraction_width => 24,
c_result_width => 32,
c_has_ce=> 1,
c_has_operation_nd => 1,
c_rate => 1,
c_latency => 3)

My problem is that now there is an error saying: "Library unit floating_point_v5_0_comp is not available in library XilinxCoreLib"

Please help me. Thanks in advance.

[meix]

Hello xsound,

If you want to use the floating point component in your design in ISE, you need to first generate the IP core using Xilinx Core Generator utility. CoreGen is an ISE accessory. In CoreGen, you can specify all the feature for the component. A bunch of VHD, VHO, XCO, NGC or EDN files will be generated for you. The VHD file shows the VHDL interface for the component. To use it, try to copy the XCO, EDN or NGC file to your ISE project folder, then add it to your ISE project.

Hope this helps.

Mei

Grettings,

I created a FIR filter in CoDeveloper and exported it to Xilinx Generic VHDL (with floating point support). When I compiled the files, the program gave me an error saying that the floating point components were invalid. I poked around and found that the new version of the floating point libraries for Xilinx is 5.0. So I changed the version to 5. This is the library declaration in my code:

library XilinxCoreLib;
use XilinxCoreLib.floating_point_v5_0_comp.all;

and this is an example of the component created by CoDeveloper:

xilinx_fadd : floating_point_v5_0
generic map(
c_has_add => 1,
c_a_width => 32,
c_a_fraction_width => 24,
c_b_width => 32,
c_b_fraction_width => 24,
c_result_fraction_width => 24,
c_result_width => 32,
c_has_ce=> 1,
c_has_operation_nd => 1,
c_rate => 1,
c_latency => 3)

My problem is that now there is an error saying: "Library unit floating_point_v5_0_comp is not available in library XilinxCoreLib"

Please help me. Thanks in advance.

[chie0017]

Hello xsound,

If you want to use the floating point component in your design in ISE, you need to first generate the IP core using Xilinx Core Generator utility. CoreGen is an ISE accessory. In CoreGen, you can specify all the feature for the component. A bunch of VHD, VHO, XCO, NGC or EDN files will be generated for you. The VHD file shows the VHDL interface for the component. To use it, try to copy the XCO, EDN or NGC file to your ISE project folder, then add it to your ISE project.

Hope this helps.

Mei

Hi, I am having a similar problem, but in my case I'm using EDK and the error message below appeared when I try to generate the bitstream:

ERROR:HDLCompiler:104 - "PATH_TO_EDK_PROJECT/EDK/pcores/impulse/hdl/vhdl/float_ll.vhd" Line 46: Cannot find <floating_point_v4_0_comp> in library <xilinxcorelib>. Please ensure that the library was compiled, and that a library and a use clause are present in the VHDL file.

The PSP I used was Xilinx MicroBlaze FSL, with "floating point library" box ticked.

If the solution is similar, can anyone tell me how to do it in XPS?

Thanks in advance!

[etrexel]

Hi,
What version of CoDeveloper are you using? and what device are you targeting? If it is a Virtex-6, Spartan-6, or newer, then due to a change in the floating point cores provided by Xilinx, you will require the latest version of CoDeveloper 3.70.b.x along with a patch available by emailing support@impulsec.com.

Best Regards,
Ed