A computing system is provided, including a processor configured to generate a directed weighted graph indicating a plurality of functions configured to be executed on a plurality of communicatively connected processing devices. For each of a plurality of pairs of the functions, the processor may determine a shortest path between the pair of functions. The processor may generate a second graph indicating the plurality of pairs of functions connected by the shortest paths. The processor may receive a pipeline directed acyclic graph (DAG) specifying a data pipeline of a plurality of processing stages. The processor may determine a subgraph isomorphism between the pipeline DAG and the second graph. The processor may convey, to one or more processing devices of the plurality of processing devices, instructions to execute the plurality of processing stages as specified by the subgraph isomorphism.

A device having a Graphics Processing Unit (GPU) may be configured to selectively run in a normal mode or a timing testing mode. In the timing testing mode the device is configured to disrupt timing of processing that takes place on the GPU while running an application with the GPU and test the application for errors in device hardware component and/or software component synchronization while the device is running in the timing testing mode.

An apparatus includes a CPU core and a L1 cache subsystem including a L1 main cache, a L1 victim cache, and a L1 controller. The apparatus includes a L2 cache subsystem including a L2 main cache, a shadow L1 main cache, a shadow L1 victim cache, and a L2 controller configured to receive a read request from the L1 controller as a single transaction. Read request includes a read address, a first indication of an address and a coherence state of a cache line A to be moved from the L1 main cache to the L1 victim cache to allocate space for data returned in response to the read request, and a second indication of an address and a coherence state of a cache line B to be removed from the L1 victim cache in response to the cache line A being moved to the L1 victim cache.

Described herein is a graphics processing unit (GPU) comprising a single instruction, multiple thread (SIMT) multiprocessor comprising an instruction cache, a shared memory coupled with the instruction cache, and circuitry coupled with the shared memory and the instruction cache, the circuitry including multiple texture units, a first core including hardware to accelerate matrix operations, and a second core configured to receive an instruction having multiple operands in a bfloat16 (BF16) number format, wherein the multiple operands include a first source operand, a second source operand, and a third source operand, and the BF16 number format is a sixteen-bit floating point format having an eight-bit exponent and process the instruction, wherein to process the instruction includes to multiply the second source operand by the third source operand and add a first source operand to a result of the multiply.

An apparatus including a CPU core and a L1 cache subsystem coupled to the CPU core. The L1 cache subsystem includes a L1 main cache, a L1 victim cache, and a L1 controller. The apparatus includes a L2 cache subsystem coupled to the L1 cache subsystem. The L2 cache subsystem includes a L2 main cache, a shadow L1 main cache, a shadow L1 victim cache, and a L2 controller. The L2 controller receives an indication from the L1 controller that a cache line A is being relocated from the L1 main cache to the L1 victim cache; in response to the indication, update the shadow L1 main cache to reflect that the cache line A is no longer located in the L1 main cache; and in response to the indication, update the shadow L1 victim cache to reflect that the cache line A is located in the L1 victim cache.

A processing apparatus comprising one or more processing modules, each comprising an execution unit. The one or more processing modules are operable to run a plurality of parallel or concurrent threads, and the processing apparatus further comprises a storage location for storing an aggregated exit state of the plurality of threads. An instruction set of the processing apparatus comprises an exit instruction for inclusion in each of the plurality of threads, the exit state instruction taking an individual exit state of the respective thread as an operand. The exit instruction terminates the respective thread and also causes the individual exit state specified in the operand to contribute to the aggregated exit state.

The present disclosure provides a computing method that is applied to a computing device. The computing device includes: a memory, a register unit, and a matrix computing unit. The method includes the following steps: controlling, by the computing device, the matrix computing unit to obtain a first operation instruction, where the first operation instruction includes a matrix reading instruction for a matrix required for executing the instruction; controlling, by the computing device, an operating unit to send a reading command to the memory according to the matrix reading instruction; and controlling, by the computing device, the operating unit to read a matrix corresponding to the matrix reading instruction in a batch reading manner, and executing the first operation instruction on the matrix. The technical solutions in the present disclosure have the advantages of fast computing speed and high efficiency.

Apparatuses, systems, and techniques to compute cyclic redundancy checks use a graphics processing unit (GPU) to compute cyclic redundancy checks. For example, in at least one embodiment, an input data sequence is distributed among GPU threads for parallel calculation of an overall CRC value for the input data sequence according to various novel techniques described herein.

Retiring instructions out-of-order includes: receiving processor instructions comprising two or more and fewer than all processor instructions generated based on a program, where the processor instructions include a first instruction and a second instruction such that the first instruction precedes the second instruction in a program order of the program; receiving a start instruction that immediately precedes the processor instructions and indicates that the processor instructions are to be retired out-of-order; receiving a stop instruction immediately that succeeds the processor instructions and indicates a stop to out-of-order instruction retirement; and, in response to completing execution of the second instruction before completing execution of the first instruction, retiring the second instruction before retiring the first instruction.

A pipeline processing unit includes a fetch unit that fetches the instruction for the thread having an execution right, a decoding unit that decodes the instruction fetched by the fetch unit, and a computation execution unit that executes the instruction decoded by the decoding unit. When the WAIT instruction for the thread having the execution right is executed, an instruction holding unit holds instruction fetch information on a processing target instruction to be processed immediately after the WAIT instruction. An execution target thread selection unit selects a thread to be executed based on a wait command and, in response to a wait state started from the execution of the WAIT instruction being canceled, processes the processing target instruction from decoding thereof based on the instruction fetch information on the processing target instruction held in the instruction holding unit.