[moved] Need VHDL code for floating point multiplier

i am doing a project on double precision floating point multiplier....can you help me to write VHDL code..???

Xilinx and altera already provide floating point ip cores.

can u give me the direct link...it helps me also alottt...plzzz..

Google can produce the results quicker than typing a post on this forum:

altera : http://www.altera.co.uk/literature/ug/ug_altfp_mfug.pdf
xilinx : http://www.xilinx.com/products/intellectual-property/floating_pt.html

in that links i get the information about floating point multiplier..but i didnt get the actual vhdl code..code is very important in my academic work..could u plz help me to find out code..

Unfortunate the floating point vendor libraries aren't provided as VHDL sources, most likely they even haven't been written in VHDL or Verilog. They are designed to use the DSP hardware of different FPGA families in an optimal way, using respective low-level primitives. Altera e.g. is typically writing this stuff in AHDL.

For the same reason, writing your own float libraries will probably not cope with the performance of the vendor libraries.

I'm not sure what's the focus of your project. If you're mainly interested in functional floating point code and don't care for performance, you can refer to the IEEE float package. It's completely missing pipelining and thus won't be used for hardware synthesis as is, but it's pure VHDL code. You can add pipelining on your own.

There are surely other projects of floating point code dedicated for hardware design. To locate it, you should start your own web search.

Another reasonable approach would be writing a floating point multiplier from the scratch.

i need only software code..i.e.,vhdl code..i am not going implement it in hardware..

VHDL is NOT software code. It is a hardware description language.

Both ALtera and xilinx provide simulation models for you to use in a simulator.