Method and apparatus for providing packed shift operations in a processor
AssigneeIntel Corporation
InventorsDerrick Chu Lin, Punit Minocha, Alexander D. Peleg, Yaakov Yaari, Millind Mittal, Larry M. Mennemeier, Benny Eitan
Life of this patent
Estimating the term…
Abstract
A method and apparatus for providing, in a processor, a shift operation on a packed data element having multiple values. One embodiment of a central processing unit (CPU) includes instruction fetch logic to fetch a single-instruction-multiple-data (SIMD) shift instruction. A register stores a multiple data elements to be operated upon by the SIMD shift instruction. A barrel shifter concurrently shifts the data elements in a bit-wise manner by a variable number of bit positions in response to the SIMD shift instruction.
Claims22
A central processing unit (CPU) comprising:
- instruction fetch logic to fetch a single-instruction-multiple-data (SIMD) shift instruction;
- at least one register to store a plurality of 16-bit data elements to be operated upon by the SIMD shift instruction;
- and
- a barrel shifter to concurrently shift the plurality of 16-bit data elements in a bit-wise manner by a variable number of bit positions in response to the SIMD shift instruction being performed.
An apparatus comprising:
- a decoder to decode a single-instruction-multiple-data (SIMD) shift instruction;
- a register to store packed data or floating point data;
- and
- an arithmetic logic unit, including a barrel shifter to concurrently shift the packed data stored in the register by a variable number of bit positions in response to the SIMD shift instruction being decoded.
A computer system comprising:
- a memory to store a single-instruction-multiple-data (SIMD) shift instruction;
- and
- a processor to fetch the SIMD shift instruction from the memory, the processor including a barrel shifter to shift two 16-bit operands identified by the SIMD shift instruction in parallel by a variable amount indicated by the SIMD shift instruction and to saturate each of the two 16-bit operands according to a saturation signal indicated by the SIMD shift instruction.
Dependencies are inferred from the claim text.