Method and apparatus for performing multiply-add operations on packed byte data

An apparatus comprising:

• a first input to receive a first packed data comprising at least four byte data elements
• a second input to receive a second packed data comprising at least four byte data elements;
• a decoder to decode a plurality of instructions including a first instruction comprising a first opcode field having a hexadecimal value of 0F38, a second opcode field having a hexadecimal value of 04, an addressing mode field, a first source field indicating a first location to access said first packed data, and a second source field indicating a second location to access said second packed data;
• a multiply-adder circuit, enabled by the decoded first instruction, to multiply each of a first pair of byte data elements of the first packed data with respective byte data elements of the second packed data and to generate a first 16-bit result representing a first sum of products of the first pair of multiplications, and to multiply each of a second pair of byte data elements of the first packed data with respective byte data elements of the second packed data and to generate a second 16-bit result representing a second sum of products of the second pair of multiplications;
• and
• an ouput to store a third packed data comprising at least said first and second 16-bit results in response to the first instruction.

A computing system comprising:

• an addressable memory to store data;
• a processor including:
• a first storage area to store M packed unsigned byte data elements;
• a second storage area to store M packed signed byte data elements;
• a decoder to decode a first instruction comprising a first opcode field having a hexadecimal value of 0F38, a second opcode field having a hexadecimal value of 04, a first source field indicating said first storage area, and a second source field indicating said second storage area;
• an execution unit, responsive to the decoder decoding a first instruction, to produce M products of multiplication of the packed byte data elements stored in the first storage area by corresponding packed byte data elements stored in the second storage area, and to sum the M products of multiplication pairwise to produce M/2 results representing M/2 sums of products;
• and
• a third storage area to store M/2 packed 16-bit data elements, the third storage area corresponding to a destination specified by the first instruction to store the M/2 results;
• and
• a magnetic storage device to store said first instruction.