Abstract: Systems and methods for an efficient and robust multiprocessor-coprocessor interface that may be used between a streaming multiprocessor and an acceleration coprocessor in a GPU are provided. According to an example implementation, in order to perform an acceleration of a particular operation using the coprocessor, the multiprocessor: issues a series of write instructions to write input data for the operation into coprocessor-accessible storage locations, issues an operation instruction to cause the coprocessor to execute the particular operation; and then issues a series of read instructions to read result data of the operation from coprocessor-accessible storage locations to multiprocessor-accessible storage locations.

Abstract: Apparatuses, systems, and techniques to adapt instructions in a SIMT architecture for execution on serial execution units. In at least one embodiment, a set of one or more threads is selected from a group of active threads associated with an instruction and the instruction is executed for the set of one or more threads on a serial execution unit.

Abstract: A technique for block data transfer is disclosed that reduces data transfer and memory access overheads and significantly reduces multiprocessor activity and energy consumption. Threads executing on a multiprocessor needing data stored in global memory can request and store the needed data in on-chip shared memory, which can be accessed by the threads multiple times. The data can be loaded from global memory and stored in shared memory using an instruction which directs the data into the shared memory without storing the data in registers and/or cache memory of the multiprocessor during the data transfer.

Abstract: Convergence of threads executing common code sections is facilitated using instructions inserted at strategic locations in computer code sections. The inserted instructions enable the threads in a warp or other group to cooperate with a thread scheduler to promote thread convergence.

Abstract: To synchronize operations of a computing system, a new type of synchronization barrier is disclosed. In one embodiment, the disclosed synchronization barrier provides for certain synchronization mechanisms such as, for example, “Arrive” and “Wait” to be split to allow for greater flexibility and efficiency in coordinating synchronization. In another embodiment, the disclosed synchronization barrier allows for hardware components such as, for example, dedicated copy or direct-memory-access (DMA) engines to be synchronized with software-based threads.

Abstract: A new synchronization system synchronizes data exchanges between producer processes and consumer processes which may be on the same or different processors in a multiprocessor system. The synchronization incurs less than one roundtrip of latency - in some implementations, in approximately 0.5 roundtrip times. A key aspect of the fast synchronization is that the producer’s data store is followed without delay with the updating of a barrier on which the consumer is waiting.

Abstract: This specification describes techniques for implementing matrix multiply and add (MMA) operations in graphics processing units (GPU)s and other processors. The implementations provide for a plurality of warps of threads to collaborate in generating the result matrix by enabling each thread to share its respective register files to be accessed by the datapaths associated with other threads in the group of warps. A state machine circuit controls a MMA execution among the warps executing on asynchronous computation units. A group MMA (GMMA) instruction provides for a descriptor to be provided as parameter where the descriptor may include information regarding size and formats of input data to be loaded into shared memory and/or the datapath.

Abstract: Distributed shared memory (DSMEM) comprises blocks of memory that are distributed or scattered across a processor (such as a GPU). Threads executing on a processing core local to one memory block are able to access a memory block local to a different processing core. In one embodiment, shared access to these DSMEM allocations distributed across a collection of processing cores is implemented by communications between the processing cores. Such distributed shared memory provides very low latency memory access for processing cores located in proximity to the memory blocks, and also provides a way for more distant processing cores to also access the memory blocks in a manner and using interconnects that do not interfere with the processing cores' access to main or global memory such as hacked by an L2 cache.

Abstract: This specification describes a programmatic multicast technique enabling one thread (for example, in a cooperative group array (CGA) on a GPU) to request data on behalf of one or more other threads (for example, executing on respective processor cores of the GPU). The multicast is supported by tracking circuitry that interfaces between multicast requests received from processor cores and the available memory. The multicast is designed to reduce cache (for example, layer 2 cache) bandwidth utilization enabling strong scaling and smaller tile sizes.

Abstract: A new level(s) of hierarchy—Cooperate Group Arrays (CGAs)—and an associated new hardware-based work distribution/execution model is described. A CGA is a grid of thread blocks (also referred to as cooperative thread arrays (CTAs)). CGAs provide co-scheduling, e.g., control over where CTAs are placed/executed in a processor (such as a GPU), relative to the memory required by an application and relative to each other. Hardware support for such CGAs guarantees concurrency and enables applications to see more data locality, reduced latency, and better synchronization between all the threads in tightly cooperating collections of CTAs programmably distributed across different (e.g., hierarchical) hardware domains or partitions.

Abstract: A parallel processing unit comprises a plurality of processors each being coupled to a memory access hardware circuitry. Each memory access hardware circuitry is configured to receive, from the coupled processor, a memory access request specifying a coordinate of a multidimensional data structure, wherein the memory access hardware circuit is one of a plurality of memory access circuitry each coupled to a respective one of the processors; and, in response to the memory access request, translate the coordinate of the multidimensional data structure into plural memory addresses for the multidimensional data structure and using the plural memory addresses, asynchronously transfer at least a portion of the multidimensional data structure for processing by at least the coupled processor. The memory locations may be in the shared memory of the coupled processor and/or an external memory.

Abstract: A new transaction barrier synchronization primitive enables executing threads and asynchronous transactions to synchronize across parallel processors. The asynchronous transactions may include transactions resulting from, for example, hardware data movement units such as direct memory units, etc. A hardware synchronization circuit may provide for the synchronization primitive to be stored in a cache memory so that barrier operations may be accelerated by the circuit. A new wait mechanism reduces software overhead associated with waiting on a barrier.

Abstract: A parallel processing unit comprises a plurality of processors each being coupled to a memory access hardware circuitry. Each memory access hardware circuitry is configured to receive, from the coupled processor, a memory access request specifying a coordinate of a multidimensional data structure, wherein the memory access hardware circuit is one of a plurality of memory access circuitry each coupled to a respective one of the processors; and, in response to the memory access request, translate the coordinate of the multidimensional data structure into plural memory addresses for the multidimensional data structure and using the plural memory addresses, asynchronously transfer at least a portion of the multidimensional data structure for processing by at least the coupled processor. The memory locations may be in the shared memory of the coupled processor and/or an external memory.

Abstract: A technique for block data transfer is disclosed that reduces data transfer and memory access overheads and significantly reduces multiprocessor activity and energy consumption. Threads executing on a multiprocessor needing data stored in global memory can request and store the needed data in on-chip shared memory, which can be accessed by the threads multiple times. The data can be loaded from global memory and stored in shared memory using an instruction which directs the data into the shared memory without storing the data in registers and/or cache memory of the multiprocessor during the data transfer.

Abstract: A technique for block data transfer is disclosed that reduces data transfer and memory access overheads and significantly reduces multiprocessor activity and energy consumption. Threads executing on a multiprocessor needing data stored in global memory can request and store the needed data in on-chip shared memory, which can be accessed by the threads multiple times. The data can be loaded from global memory and stored in shared memory using an instruction which directs the data into the shared memory without storing the data in registers and/or cache memory of the multiprocessor during the data transfer.

Abstract: Convergence of threads executing common code sections is facilitated using instructions inserted at strategic locations in computer code sections. The inserted instructions enable the threads in a warp or other group to cooperate with a thread scheduler to promote thread convergence.

Abstract: Convergence of threads executing common code sections is facilitated using instructions inserted at strategic locations in computer code sections. The inserted instructions enable the threads in a warp or other group to cooperate with a thread scheduler to promote thread convergence.

Abstract: Approaches in accordance with various embodiments provide for the processing of sparse matrices for mathematical and programmatic operations. In particular, various embodiments enforce sparsity constraints for performing sparse matrix multiply-add instruction (MMA) operations. Deep neural networks can exhibit significant sparsity in the data used in operations, both in the activations and weights. The computational load can be reduced by excluding zero-valued data elements. A sparsity constraint is applied across all submatrices of a sparse matrix, providing fine-grained structured sparsity that is evenly distributed across the matrix. The matrix may then be compressed since a minimum number of elements of the matrix are known to have zero value. Matrix operations are then performed using these matrices.

Abstract: A parallel processing unit (e.g., a GPU), in some examples, includes a hardware scheduler and hardware arbiter that launch graphics and compute work for simultaneous execution on a SIMD/SIMT processing unit. Each processing unit (e.g., a streaming multiprocessor) of the parallel processing unit operates in a graphics-greedy mode or a compute-greedy mode at respective times. The hardware arbiter, in response to a result of a comparison of at least one monitored performance or utilization metric to a user-configured threshold, can selectively cause the processing unit to run one or more compute work items from a compute queue when the processing unit is operating in the graphics-greedy mode, and cause the processing unit to run one or more graphics work items from a graphics queue when the processing unit is operating in the compute-greedy mode. Associated methods and systems are also described.

Abstract: A unified cache subsystem includes a data memory configured as both a shared memory and a local cache memory. The unified cache subsystem processes different types of memory transactions using different data pathways. To process memory transactions that target shared memory, the unified cache subsystem includes a direct pathway to the data memory. To process memory transactions that do not target shared memory, the unified cache subsystem includes a tag processing pipeline configured to identify cache hits and cache misses. When the tag processing pipeline identifies a cache hit for a given memory transaction, the transaction is rerouted to the direct pathway to data memory. When the tag processing pipeline identifies a cache miss for a given memory transaction, the transaction is pushed into a first-in first-out (FIFO) until miss data is returned from external memory. The tag processing pipeline is also configured to process texture-oriented memory transactions.