Residual Addition for Video Software Techniques
AssigneeIntel Corporation
InventorsBradley C. Aldrich, Nigel C. Paver, Murli Ganeshan
Life of this patent
Estimating the term…
Abstract
According to some embodiments, a technique provides for the execution of an instruction that includes receiving residual data of a first image and decoded pixels of a second image, zero-extending a plurality of unsigned data operands of the decoded pixels producing a plurality of unpacked data operands, adding a plurality of signed data operands of the residual data to the plurality of unpacked data operands producing a plurality of signed results; and saturating the plurality of signed results producing a plurality of unsigned results.
Claims21
A method of executing an instruction comprising:
- receiving residual data of a first image and decoded pixels of a second image;
- zero-extending a plurality of unsigned data operands of the decoded pixels producing a plurality of unpacked data operands;
- adding a plurality of signed data operands of the residual data to the plurality of unpacked data operands producing a plurality of signed results;
- saturating the plurality of signed results producing a plurality of unsigned results.
A method comprising:
- decoding an instruction identifying a mixed-mode addition operation;
- executing the instruction on a first source and a second source, wherein the first source comprises a plurality of signed residual data of a first image and the second source comprises a plurality of unsigned decoded pixels of a second image;
- and
- storing an output of the executing the instruction, wherein the output comprises a plurality of unsigned result pixels;
- wherein the executing the instruction comprises:
- zero-extending at least one of the plurality of unsigned decoded pixels;
- adding the at least one of the plurality of unsigned decoded pixels and the plurality of signed residual data producing a plurality of signed sums;
- and
- saturating the plurality of signed sums producing the plurality of unsigned result pixels.
An apparatus comprising:
- a first plurality of multiplexers, each multiplexer of the first plurality of multiplexers operative to select one of a plurality of unsigned decoded pixels and zero-extend the unsigned decoded pixels, the first plurality of multiplexers operative to produce a plurality of unpacked operands;
- a plurality of adders, each adder of the plurality of adders operative to add a signed residual data operand to one of the plurality of unpacked operands, the plurality of adders operative to produce a plurality of sums,
- a plurality of saturation units operative to produce a plurality of unsigned result pixels from the plurality of sums.
-36. (canceled)
Dependencies are inferred from the claim text.