Abstract: In an example, an apparatus comprises a plurality of execution units, and logic, at least partially including hardware logic, to assemble a general register file (GRF) message and hold the GRF message in storage in a data port until all data for the GRF message is received. Other embodiments are also disclosed and claimed.

Abstract: Embodiments are generally directed to providing power savings for a neural network architecture with zero activations during inference. An embodiment of an apparatus includes one or more processors including one or more processor cores; and a memory to store data for processing including neural network processing, wherein the apparatus to perform a fast clear operation to initialize activation buffers for a neural network by updating metadata to indicate zero values, the neural network including a plurality of layers, wherein the apparatus is to compare outputs for the neural network to the metadata values and to write an output to memory only if the output is non-zero.

Abstract: A mechanism is described for facilitating fabric-based compression and/or decompression of data at computing devices. A method of embodiments, as described herein, includes compressing contents of a data stream traveling through an internal fabric between a source component and a destination component, wherein the contents are compressed on the internal fabric.

Abstract: Systems, apparatuses and methods may a performance-enhanced computing system comprising a sensor for measuring luminance values corresponding to light focused onto the sensor at a plurality of pixel locations, a memory including a set of instructions, and a processor. The processor executes a set of instructions causing the system to generate a multi-segment tone mapping curve, generate a set of tone mapping values corresponding to the multi-segment tone mapping curve for equally spaced input values between zero and one for storage into a look up table, and process the luminance values using the look up table to apply the tone mapping curve to the luminance values of the pixels.

Abstract: The systems, apparatuses and methods may provide a way to adaptively process and aggressively cull geometry data. Systems, apparatuses and methods may provide for processing, by a positional only shading pipeline (POSH), geometry data including surface triangles for a digital representation of a scene. More particularly, systems, apparatuses and methods may provide a way to identify surface triangles in one or more exclusion zones and non-exclusion zones, and cull surface triangles surface triangles in one or more exclusion zones.

Abstract: In an example, an apparatus comprises a plurality of execution units, and a first memory communicatively couple to the plurality of execution units, wherein the first shared memory is shared by the plurality of execution units and a copy engine to copy context state data from at least a first of the plurality of execution units to the first shared memory. Other embodiments are also disclosed and claimed.

Abstract: An apparatus to facilitate data intelligent dispatching is disclosed. The apparatus includes one or more processing units including a plurality of execution units (EUs) to execute a plurality of processing threads and collection logic to collect statistics data for threads executed at the processing unit during execution of an application, and dispatch logic to dispatch the threads to be executed at a subset of the plurality of EUs during a subsequent execution of the application based on the statistics data.

Abstract: One embodiment provides for a general-purpose graphics processing unit comprising a processing array including multiple compute blocks, each compute block including multiple processing clusters and a thread dispatch unit to dispatch threads of a workload to the multiple compute blocks based on a parallelism metric, wherein the thread dispatch unit, based on the parallelism metric, is to perform one of a first operation and a second operation, the first operation to distribute threads across the multiple compute blocks and the second operation is to concentrate threads within one of the multiple compute blocks.

Abstract: Embodiments described herein provide a general purpose graphics processor comprising a plurality of tiles, each tile of the plurality of tiles comprising at least one execution unit, a local cache, and a cache control unit, and a high bandwidth memory communicatively coupled to the plurality of tiles, wherein the high bandwidth memory is shared between the plurality of tiles. The cache control unit is to implement a partial write management protocol to receive a partial write operation directed to a cache line in the local cache, the partial write operation comprising write data, write the data associated with the partial write operation to the local cache when the cache line is in a modified state, and forward the write data associated with the partial write operation to the high bandwidth memory when the partial write operation triggers a cache miss or when the cache line is in an exclusive state or a shared state. Other embodiments may be described and claimed.

Abstract: Systems, apparatuses and methods may provide away to render edges of an object defined by multiple tessellation triangles. More particularly, systems, apparatuses and methods may provide a way to perform anti-aliasing at the edges of the object based on a coarse pixel rate, where the coarse pixels may be based on a coarse Z value indicate a resolution or granularity of detail of the coarse pixel. The systems, apparatuses and methods may use a shader dispatch engine to dispatch raster rules to a pixel shader to direct the pixel shader to include, in a tile and/or tessellation triangle, one more finer coarse pixels based on a percent of coverage provided by a finer coarse pixel of a tessellation triangle at or along the edge of the object.

Abstract: Embodiments described herein provide an apparatus comprising a processor to allocate a first memory space for data for a graphics workload, the first memory comprising a first plurality of addressable memory locations, allocate a second memory space for compression metadata relating to the data for the graphics workload, the second memory space comprising a second plurality of addressable memory locations and having an amount of memory corresponding to a predetermined ratio of the amount of memory allocated to first memory space, and configure a direct memory mapping between the first plurality of addressable memory locations and the second plurality of addressable memory locations. Other embodiments may be described and claimed.

Abstract: One embodiment provides for a general-purpose graphics processor comprising a hardware graphics rendering pipeline configured to perform multisample anti-aliasing, the hardware graphics rendering pipeline including a pixel processing unit configured to generate pixel color data in a graphics processing pipeline, the pixel processing unit to output color data to a multisample render target, the multisample render target to store multiple sample locations for each pixel in a set of pixels. The general-purpose graphics processor further comprises a memory allocator to allocate memory to store color data associated with the multisample render target, the memory allocator to merge a memory allocation for multiple pixels having a sample associated with a same color value.

Abstract: The invention relates to A frozen confection coating composition, the composition comprising, expressed in weight % based on the total weight of the coating, 35-75 wt. % of non-interesterified fat, preferably 40-65 wt. % of non-interesterified fat which comprises a fat blend of medium soft fat and liquid oil, and 25-65 wt. % of non-fat solids, preferably 35-60 wt. % of non-fat solids, wherein, the coating composition comprises, less than 35 wt. % of saturated fatty acid, preferably less than 30 wt. % of saturated fatty acids 15-50 wt. %, preferably 18-30% of monounsaturated fatty acid and less than 10%, preferably less than 5% of polyunsaturated fatty acid, and wherein the medium soft fat has above 40%, preferably between 50-70%, of solid fat content at 20° C., and medium soft fat has 54-60% of saturated fatty acid, and wherein the fat blend in the coating crystallizes in a first and second crystallization step at a temperature below ?15° C. and displays a solid fat content of 30-50% within 2 min.

Abstract: An apparatus and method for protecting content in a graphics processor. For example, one embodiment of an apparatus comprises: encode/decode circuitry to decode protected audio and/or video content to generate decoded audio and/or video content; a graphics cache of a graphics processing unit (GPU) to store the decoded audio and/or video content; first protection circuitry to set a protection attribute for each cache line containing the decoded audio and/or video data in the graphics cache; a cache coherency controller to generate a coherent read request to the graphics cache; second protection circuitry to read the protection attribute to determine whether the cache line identified in the read request is protected, wherein if it is protected, the second protection circuitry to refrain from including at least some of the data from the cache line in a response.

Abstract: A mechanism is described for facilitating using of a shared local memory for register spilling/filling relating to graphics processors at computing devices. A method of embodiments, as described herein, includes reserving one or more spaces of a shared local memory (SLM) to perform one or more of spilling and filling relating to registers associated with a graphics processor of a computing device.

Abstract: A mechanism is described for facilitating dynamic merging of atomic operations in computing devices. A method of embodiments, as described herein, includes facilitating detecting atomic messages and a plurality of slot addresses. The method further includes comparing one or more slot addresses of the plurality of slot addresses with other slot addresses of the plurality of slot addresses to seek one or more matched slot addresses, where the one or more matched slot addresses are merged into one or more merged groups. The method may further include generating one or more merged atomic operations based on and corresponding to the one or more merged groups.

Abstract: Examples are described here that can be used to allocate commands from multiple sources to performance by one or more segments of a processing device. For example, a processing device can be segmented into multiple portions and each portion is allocated to process commands from a particular source. In the event a single source provides commands, the entire processing device (all segments) can be allocated to process commands from the single source. When a second source provides commands, some segments can be allocated to perform commands from the first source and other segments can be allocated to perform commands from the second source. Accordingly, commands from multiple applications can be executed by a processing unit at the same time.

Abstract: Embodiments described herein provide an apparatus comprising a processor to allocate a first memory space for data for a graphics workload, the first memory comprising a first plurality of addressable memory locations, allocate a second memory space for compression metadata relating to the data for the graphics workload, the second memory space comprising a second plurality of addressable memory locations and having an amount of memory corresponding to a predetermined ratio of the amount of memory allocated to first memory space, and configure a direct memory mapping between the first plurality of addressable memory locations and the second plurality of addressable memory locations. Other embodiments may be described and claimed.

Abstract: An apparatus to facilitate memory data compression is disclosed. The apparatus includes a memory and having a plurality of banks to store main data and metadata associated with the main data and a memory management unit (MMU) coupled to the plurality of banks to perform a hash function to compute indices into virtual address locations in memory for the main data and the metadata and adjust the metadata virtual address locations to store each adjusted metadata virtual address location in a bank storing the associated main data.

Abstract: In an example, an apparatus comprises a plurality of execution units, and a cache memory communicatively coupled to the plurality of execution units, wherein the cache memory is structured into a plurality of sectors, wherein each sector in the plurality of sectors comprises at least two cache lines. Other embodiments are also disclosed and claimed.