摘要:
Methods and apparatuses for transposing a matrix using a vector look up unit. In one aspect of the invention, a method for matrix transposition includes: rotating in a vector register a first row of a matrix to generate a first row, of elements; writing simultaneously into a plurality of look up units the first row of elements indexed by a first row of indices in a vector register; looking up simultaneously from the plurality of look up units a second row of elements indexed by a second row of indices in a vector register; and rotating in a vector register the second row of elements to generate a third row of elements.
摘要:
Methods and apparatuses for blending two images using vector table look up operations. In one aspect of the invention, a method to blend two images includes: loading a vector of keys into a vector register; converting the vector of keys into a first vector of blending factors for the first image and a second vector of blending factors for the second image using a plurality of look up tables; and computing an image attribute for the blended image using the blending factors.
摘要:
Methods and apparatuses for looking up vectors in parallel using vector table look up operations. In one aspect of the invention, a method to look up a plurality of data items indexed by a vector of indices includes: generating a second vector of indices in a vector register where each index of the second vector of indices is one of a first vector of indices and at least one index in the first vector of indices is replicated as a plurality of duplicated indices in the second vector of indices; and looking up simultaneously a first vector of data items from a plurality of look up tables using the second vector of indices.
摘要:
Methods and apparatuses for converting color components in one space to those in another space. In one aspect of the invention, a method for vector conversion including: loading a first vector of components into a vector register; generating a first vector of indices in a vector register, each index of the first vector of indices being one of the first vector of components, at least one component in the first vector of components being replicated as a plurality of duplicated indices in the first vector of indices; looking up simultaneously a first vector of data items from a plurality of look up tables using the first vector of indices; and summing up at least one subset of the first vector of data items into at least one component of a second vector of components.
摘要:
Methods and apparatuses for blending two images using vector table look up operations. In one aspect of the invention, a method to blend two images includes: loading a vector of keys into a vector register; converting the vector of keys into a first vector of blending factors for the first image and a second vector of blending factors for the second image using a plurality of look up tables; and computing an image attribute for the blended image using the blending factors.
摘要:
Methods and apparatuses for blending two images using vector table look up operations. In one aspect of the invention, a method to blend two images includes: loading a vector of keys into a vector register; converting the vector of keys into a first vector of blending factors for the first image and a second vector of blending factors for the second image using a plurality of look up tables; and computing an image attribute for the blended image using the blending factors.
摘要:
At least one chip of a chipset in a computer system having at least one host processor and a host memory are described herein. In one aspect of the invention, an exemplary chip includes an interconnect, a memory interface coupled to the interconnect, the memory interface providing access to the host memory and controlling memory refresh and memory access, a host interface coupled to the interconnect, the host interface providing access to the host processor, and a programmable media processor coupled to the interconnect, the media processor accessing the host through the host interface and the media processor accessing the host memory through the memory interface, wherein the media processor processes time based media.
摘要:
An apparatus, method, and computer program product for scaling a rendered surface. The apparatus includes a graphics processor configured to render a surface at a source resolution, wherein a scaling parameter is associated with the surface, the scaling parameter defining the source resolution and a destination resolution. The apparatus further includes a display processor configured to scale the rendered surface to the destination resolution. Preferably, the scaling parameters are EGL surface attributes.
摘要:
An apparatus, method, and computer program product for scaling a rendered surface. The apparatus includes a graphics processor configured to render a surface at a source resolution, wherein a scaling parameter is associated with the surface, the scaling parameter defining the source resolution and a destination resolution. The apparatus further includes a display processor configured to scale the rendered surface to the destination resolution. Preferably, the scaling parameters are EGL surface attributes.
摘要:
In general, the present disclosure describes various techniques for programmable, pattern-based unpacking and packing of data channel information, including still image, video, and audio component data. One example device comprises a programmable processor having a plurality of processing pipelines. The processor is configured to receive pattern information that specifies a pattern for a plurality of input data components, the pattern information comprising a plurality of pattern elements that are each associated with one or more of the input data components, and each input data component being selected from a component group consisting of a still image data component, an audio data component, and a video data component. For example, the input data components may comprise pixel data components, such as color channels. The processor is further configured to provide each input data component to a selected processing pipeline of the processor in accordance with the pattern information.