Building Diamond applications

Diamond uses a three-stage approach to building an application from C source files (FPGA components are built with the Xilinx tools):

Compile your source files with the 3L C command. This invokes the standard Texas Instruments compiler or assembler.

Use the 3L T command, usually several times, to generate a number of separate task files. This invokes the TI linker.

Dragons003 You do not usually need linker command files and, even in the rare cases when you do use them, you never use the MEMORY or SECTIONS directives (although there is one use of SECTIONS in Diamond required to work round a problem introduced by unfortunate changes in the Texas Instruments linker, but this should be invisible to most users).

Create a configuration file and then use the 3L A command to invoke the Diamond configurer to combine your task files into a single application file containing everything needed to get your application running on a network of processors. Often you do not need to worry about what memory will be allocated to your application; this is done automatically for you by the configurer, although you can exercise more control when necessary.

Once a Diamond application has been constructed, it is loaded into your processors and executed by a server program running on the PC, invoked with the 3L X command.

A Diamond application file is not a COFF (or ELF) file and cannot be loaded using Code Composer. Formats like COFF & ELF do not have the constructs necessary to support the sophisticated multiprocessor loading techniques required by an advanced system such as Diamond, nor do they have facilities for loading FPGA bitstreams.

To get a feel for how you work with Diamond you should follow the confidence test described earlier or view the short tutorial video on the 3L web site.

The various steps needed to build an application can be performed automatically by the Diamond IDE.