INSTRUCTIONS TO CONVERT FROM FP16 TO BF8

Organization Name

Intel Corporation

Inventor(s)

Alexander Heinecke of San Jose CA US

Naveen Mellempudi of Bangalore IN

Robert Valentine of Kiryat Tivon IL

Mark Charney of Lexington MA US

Christopher Hughes of Santa Clara CA US

Evangelos Georganas of San Mateo CA US

Zeev Sperber of Zichron Yackov IL

Amit Gradstein of Binyamina IL

Simon Rubanovich of Haifa IL

INSTRUCTIONS TO CONVERT FROM FP16 TO BF8

This abstract first appeared for US patent application 18927097 titled 'INSTRUCTIONS TO CONVERT FROM FP16 TO BF8

Original Abstract Submitted

Techniques for converting FP16 to BF8 using bias are described. An exemplary embodiment utilizes decoder circuitry to decode a single instruction, the single instruction to include one or more fields to identify a first source operand, one or more fields to identify a second source operand, one or more fields to identify a source/destination operand, and one or more fields for an opcode, wherein the opcode is to indicate that execution circuitry is to convert packed half-precision data from the identified first and second sources to packed bfloat8 data using bias terms from the identified source/destination operand and store the packed bfloat8 data into corresponding data element positions of the identified source/destination operand; and execution circuitry to execute the decoded instruction according to the opcode to convert packed half-precision data from the identified first and second sources to packed bfloat8 data using bias terms from the identified source/destination operand and store the packed bfloat8 data into corresponding data element positions of the identified source/destination operand.