Method and apparatus for parallel conversion of color values from a single precision floating point format to an integer format
AssigneeIntel Corporation
InventorsHsien-Cheng E. Hsieh, Thomas R. Huff, Vladimir Pentkovski, Patrice Roussel, Shreekant S. Thakkar
Life of this patent
Estimating the term…
Abstract
A method and apparatus for parallel processing of graphics data are described. A number of color components are stored in a floating point format in at least one register of a set of 128-bit registers in a packed format. The color components in the floating point format are converted to numbers in an integer format. The numbers in the integer format are placed in at least one register of a set of 64-bit registers in the packed format. Color components are assembled for image pixels from the numbers in the integer format.
Claims49
A method for parallel processing of graphics data, the method comprising the steps of:
- storing a plurality of color components in a floating point format in at least one register of a set of 128-bit registers in the packed format;
- converting the plurality of color components in a floating point format to a plurality of numbers in an integer format;
- placing the plurality of numbers in the integer format in at least one register of a set of 64-bit registers in the packed format;
- and
- assembling color components for each of a plurality of pixels from the plurality of numbers in the integer format in the at least one register of a set of 64-bit registers.
At least one instruction for parallel processing of graphics data, which when executed causes a computer to perform the following steps:
- storing a plurality of color components in a floating point format in at least one register of a set of 128-bit registers in the packed format;
- converting the plurality of color components in a floating point format to a plurality of numbers in an integer format;
- placing the plurality of numbers in the integer format in at least one register of a set of 64-bit registers in the packed format;
- and
- assembling color components for each of a plurality of pixels from the plurality of numbers in the integer format in the at least one register of a set of 64-bit registers.
A processor comprising:
- a decoder configured to decode instructions, and
- a circuit coupled to the decoder, the circuit in response to at least one decoded instruction being configured to,
- store a plurality of color components in a floating point format in at least one register of a set of 128-bit registers in the packed format,
- convert the plurality of color components in a floating point format to a plurality of numbers in an integer format,
- place the plurality of numbers in the integer format in at least one register of a set of 64-bit registers in the packed format, and
- assemble color components for each of a plurality of pixels from the plurality of numbers in the integer format in the at least one register of a set of 64-bit registers.
Dependencies are inferred from the claim text.