Method for performing multiply-add operations on packed data

In a computer system, a method comprising:

• responsive to the execution of a single instruction that specifies a first packed data and a second packed data, said first packed data including A1, A2, A3, A4 as data elements, said second packed data including B1, B2, B3, and B4 as data elements, performing
• performing the operation (A1×B1)+(A2×B2) to generate a first data element in a third packed data;
• performing the operation (A3×B3)+(A4×B 4) to generate a second data element in said third packed data;
• completing execution of said single instruction without adding said first and second data elements of said third packed data;
• and
• storing said third packed data for use as an operand to another instruction.

In a computer system having stored therein a first packed data and a second packed data each containing initial data elements, each of said initial data elements in said first packed data having a corresponding initial data element in said second packed data, a method for performing multiply add operations in response to a single instruction, said method comprising the steps of:

• multiplying together said corresponding initial data elements in said first packed data and said second packed data to generate corresponding intermediate data elements, said intermediate data elements being divided into a number of sets;
• generating a plurality of result data elements, a first of said plurality of result data elements representing the sum of said intermediate result data elements in a first of said number of sets, a second of said plurality of result data elements representing the sum of said intermediate result data elements in a second of said number of sets;
• and
• completing execution of said single instruction without summing said plurality of result data elements.

In a computer system, a method comprising:

• responsive to the execution of a single instruction that specifies a first packed data including A1, A2, A3, and A4 as data elements, and a second packed data including B1, B2, B3, and B4 as data elements, performing,
• multiplying together A1 and B1 to generate a first intermediate result;
• multiplying together A2 and B2 to generate a second intermediate result;
• multiplying together A3 and B3 to generate a third intermediate result;
• multiplying together A4 and B4 to generate a fourth intermediate result;
• performing in parallel the following steps:
• adding together said first intermediate result and said second intermediate result to generate a first data element in a third packed data;
• and
• adding together said third intermediate result and said fourth intermediate result to generate a second data element in said third packed data;
• and
• saving said third packed data for use as an operand to another instruction.

In a computer system, a method comprising:

• responsive to the execution of a single instruction that specifies a first packed data and a second packed data each containing initial data elements, each of said initial data elements in said first packed data having a corresponding initial data element in said second packed data, performing
• multiplying together said corresponding initial data elements in said first packed data and said second packed data to generate corresponding intermediate data elements, said intermediate data elements being divided into a number of sets;
• generating a plurality of result data elements, a first of said plurality of result data elements representing the sum of said intermediate result data elements in a first of said number of sets, a second of said plurality of result data elements representing the sum of said intermediate result data elements in a second of said number of sets;
• and
• completing execution of said single instruction without summing said plurality of result data elements.

A computer-implemented method comprising:

• responsive to the execution of a single instruction that specifies a first and second storage areas having respectively stored therein a first and second packed data, said first and second packed data each having a first, second, third, and fourth data elements, performing,
• multiplying together said first data elements to generate a first intermediate result,
• multiplying together said second data elements to generate a second intermediate result,
• multiplying together said third data elements to generate a third intermediate result,
• multiplying together said fourth data elements to generate a fourth intermediate result,
• adding together said first intermediate result and said second intermediate result to generate a fifth intermediate result,
• adding together said third intermediate result and said fourth intermediate result to generate a sixth intermediate result, and
• storing said fifth and sixth intermediate results as first and second unaccumulated data elements of a third packed data, respectively, wherein said third packed data includes only said first and second data elements.

In a computer system, a method comprising:

• responsive to the execution of a single instruction, performing,
• fetching a first packed data and a second packed data, said first packed data including A1, A2, A3, and A4 as data elements, said second packed data including B1, B2, B3, B4, as data elements;
• performing the operation (A1×B1)+(A2×B2) to generate a first result;
• performing the operation (A3×B3)+(A4×B4) to generate a second result;
• storing said first and second results as unaccumulated data elements in a third storage area, said third storage area having at least a first field and a second field, said first field for saving said first result as a first unaccumulated data element of said third storage area, and said second field for saving said second result as a second unaccumulated data element of said third storage area.