contains four tables which contain information about the DLX instruction set. The first is opTable. This table contains 64 entries corresponding to the 64 possible values of the opcode field. Each entry consists of an instruction-format pair. For example, the value of opTable[5] is {OP_BNEZ, IFMT} indicating that opcode 5 is a branch not equal to zero instruction, which uses the I-type format. Several entries in this table have OP_RES as the instruction. These entries are reserved for future extensions to the DLX instruction set.
The zero opcode indicates a different table should be used to identify the instruction. A second table called specialTable handles this case. In this table are all the register-register operations. The format is not specified explicitly for these instructions (as it was in opTable) because they are all R-type format. These instructions all contain a zero in the opcode field and a function encoding in the lower six bits of the instruction word. There is also room in this table for expansion by using entries currently containing OP_RES.
An opcode of one indicates a floating point arithmetic operation. A third table, FParithTable handles these instructions. As with specialTable, all instructions in this table have R-type format. The exact operation is again specified by the lower six bits of the instruction word, which are used to index into this table. Currently 32 entries contain OP_RES and are available for future expansion to the floating point instruction set.
The final table is operationNames. This table contains a list of all the integer instruction names followed by the floating point instruction names. Each group is arranged alphabetically. These tables are used to print out the names of the instructions when a dynamic instruction count is requested.