Abstract: Embodiments described herein include software, firmware, and hardware that provides techniques to enable deterministic scheduling across multiple general-purpose graphics processing units. One embodiment provides a multi-GPU architecture with uniform latency. One embodiment provides techniques to distribute memory output based on memory chip thermals. One embodiment provides techniques to enable thermally aware workload scheduling. One embodiment provides techniques to enable end to end contracts for workload scheduling on multiple GPUs.

Abstract: One embodiment provides an apparatus comprising an interconnect fabric comprising a processing cluster including an array of multiprocessors coupled to an interconnect fabric, scheduling circuitry to distribute a plurality of thread groups across the array of multiprocessors, each thread group comprising a plurality of threads. A first multiprocessor of the array of multiprocessors can be assigned to process a first thread group comprising a first plurality of threads including a first thread sub-group and a second thread sub-group. The second thread sub-group has a data dependency on the first thread sub-group and the first multiprocessor includes circuitry to cause threads of the second thread sub-group to sleep until the threads of the first thread sub-group have satisfied the data dependency.

Abstract: Apparatus and method for stack throttling.

Abstract: One embodiment provides a graphics processor comprising memory access circuitry configured to receive a message from an instruction execution resource and determine a destination for the message, the destination one of shared function circuitry of a graphics core or a set of memory banks within the graphics core. The memory access circuitry then routes the message to the shared function circuitry in response to a determination that the message is directed to the shared function circuitry or routes the message to a message sequencer associated with the instruction execution resource in response to a determination that the message is directed to the set of memory banks.

Abstract: A mechanism is described for facilitating barriers and synchronization for machine learning at autonomous machines. A method of embodiments, as described herein, includes detecting thread groups relating to machine learning associated with one or more processing devices. The method may further include facilitating barrier synchronization of the thread groups across multiple dies such that each thread in a thread group is scheduled across a set of compute elements associated with the multiple dies, where each die represents a processing device of the one or more processing devices, the processing device including a graphics processor.

Abstract: Embodiments are generally directed to cache structure and utilization. An embodiment of an apparatus includes one or more processors including a graphics processor; a memory for storage of data for processing by the one or more processors; and a cache to cache data from the memory; wherein the apparatus is to provide for dynamic overfetching of cache lines for the cache, including receiving a read request and accessing the cache for the requested data, and upon a miss in the cache, overfetching data from memory or a higher level cache in addition to fetching the requested data, wherein the overfetching of data is based at least in part on a current overfetch boundary, and provides for data is to be prefetched extending to the current overfetch boundary.

Abstract: One embodiment provides a parallel processor comprising a memory interface and a processing array coupled with the memory interface. The processing array is configured to address memory accessed via the memory interface via a virtual address mapping and includes circuitry to resolve a page fault for the virtual address mapping, wherein each of the multiple compute blocks is separately preemptable.

Abstract: In accordance with some embodiments, the render rate is varied across and/or up and down the display screen. This may be done based on where the user is looking in order to reduce power consumption and/or increase performance. Specifically the screen display is separated into regions, such as quadrants. Each of these regions is rendered at a rate determined by at least one of what the user is currently looking at, what the user has looked at in the past and/or what it is predicted that the user will look at next. Areas of less focus may be rendered at a lower rate, reducing power consumption in some embodiments.

Abstract: In one embodiment, a processor includes: a graphics processor to execute a workload; and a power controller coupled to the graphics processor. The power controller may include a voltage ramp circuit to receive a request for the graphics processor to operate at a first performance state having a first operating voltage and a first operating frequency and cause an output voltage of a voltage regulator to increase to the first operating voltage. The voltage ramp circuit may be configured to enable the graphics processor to execute the workload at an interim performance state having an interim operating voltage and an interim operating frequency when the output voltage reaches a minimum operating voltage. Other embodiments are described and claimed.

Abstract: An apparatus to facilitate compute optimization is disclosed. The apparatus includes a mixed precision core including mixed-precision execution circuitry to execute one or more of the mixed-precision instructions to perform a mixed-precision dot-product operation comprising to perform a set of multiply and accumulate operations.

Abstract: Embodiments are generally directed to compression in machine learning and deep learning processing. An embodiment of an apparatus for compression of untyped data includes a graphical processing unit (GPU) including a data compression pipeline, the data compression pipeline including a data port coupled with one or more shader cores, wherein the data port is to allow transfer of untyped data without format conversion, and a 3D compression/decompression unit to provide for compression of untyped data to be stored to a memory subsystem and decompression of untyped data from the memory subsystem.

Abstract: Systems and methods for cache utilization are disclosed. In one embodiment, a graphics processor includes processing resources to perform graphics operations and a cache controller of a cache memory that is coupled to the processing resources. The cache controller is configured to set an initial aging policy using an aging field based on age of cache lines within the cache memory and to determine whether a hint or an instruction to indicate a level of aging has been received. In one embodiment, the cache memory configured to be partitioned into multiple cache regions, wherein the multiple cache regions include a first cache region having a cache eviction policy with a configurable level of data persistence.

Abstract: Embodiments are generally directed to graphics processor data access and sharing. An embodiment of an apparatus includes a circuit element to produce a result in processing of an application; a load-store unit to receive the result and generate pre-fetch information for a cache utilizing the result; and a prefetch generator to produce prefetch addresses based at least in part on the pre-fetch information; wherein the load-store unit is to receive software assistance for prefetching, and wherein generation of the pre-fetch information is based at least in part on the software assistance.

Abstract: Embodiments provide mechanisms to facilitate compute operations for deep neural networks. One embodiment comprises a graphics processing unit comprising one or more multiprocessors, at least one of the one or more multiprocessors including a register file to store a plurality of different types of operands and a plurality of processing cores. The plurality of processing cores includes a first set of processing cores of a first type and a second set of processing cores of a second type. The first set of processing cores are associated with a first memory channel and the second set of processing cores are associated with a second memory channel.

Abstract: Embodiments are generally directed to data prefetching for graphics data processing. An embodiment of an apparatus includes one or more processors including one or more graphics processing units (GPUs); and a plurality of caches to provide storage for the one or more GPUs, the plurality of caches including at least an L1 cache and an L3 cache, wherein the apparatus to provide intelligent prefetching of data by a prefetcher of a first GPU of the one or more GPUs including measuring a hit rate for the Li cache; upon determining that the hit rate for the L1 cache is equal to or greater than a threshold value, limiting a prefetch of data to storage in the L3 cache, and upon determining that the hit rate for the L1 cache is less than a threshold value, allowing the prefetch of data to the L1 cache.

Abstract: A library of machine learning primitives is provided to optimize a machine learning model to improve the efficiency of inference operations. In one embodiment a trained convolutional neural network (CNN) model is processed into a trained CNN model via pruning, convolution window optimization, and quantization.

Abstract: Described herein are various embodiments of reducing dynamic power consumption within a processor device. One embodiment provides a technique for dynamic link width adjustment based on throughput demand for client of an interconnect fabric. One embodiment provides for a parallel processor comprising an interconnect fabric including a dynamically configurable bus widths and frequencies.

Abstract: One embodiment provides for a graphics processing unit to accelerate machine-learning operations, the graphics processing unit comprising a multiprocessor having a single instruction, multiple thread (SIMT) architecture, the multiprocessor to execute at least one single instruction; and a first compute unit included within the multiprocessor, the at least one single instruction to cause the first compute unit to perform a two-dimensional matrix multiply and accumulate operation, wherein to perform the two-dimensional matrix multiply and accumulate operation includes to compute an intermediate product of 16-bit operands and to compute a 32-bit sum based on the intermediate product.

Abstract: An apparatus to facilitate machine learning matrix processing is disclosed. The apparatus comprises a memory to store matrix data one or more processors to execute an instruction to examine a message descriptor included in the instruction to determine a type of matrix layout manipulation operation that is to be executed, examine a message header included in the instruction having a plurality of parameters that define a two-dimensional (2D) memory surface that is to be retrieved, retrieve one or more blocks of the matrix data from the memory based on the plurality of parameters and a register file including a plurality of registers, wherein the one or more blocks of the matrix data is stored within a first set of the plurality of registers.

Abstract: One embodiment provides circuitry coupled with cache memory and a memory interface, the circuitry to compress compute data at multiple cache line granularity, and a processing resource coupled with the memory interface and the cache memory. The processing resource is configured to perform a general-purpose compute operation on compute data associated with multiple cache lines of the cache memory. The circuitry is configured to compress the compute data before a write of the compute data via the memory interface to the memory bus, in association with a read of the compute data associated with the multiple cache lines via the memory interface, decompress the compute data, and provide the decompressed compute data to the processing resource.