Abstract: A processor uses the same virtual address space for heterogeneous processing units of the processor. The processor employs different sets of page tables for different types of processing units, such as a CPU and a GPU, wherein a memory management unit uses each set of page tables to translate virtual addresses of the virtual address space to corresponding physical addresses of memory modules associated with the processor. As data is migrated between memory modules, the physical addresses in the page tables can be updated to reflect the physical location of the data for each processing unit.

Abstract: Resources, such as machine learning models, can be updated for an application without any significant downtime for that application. For an application hosted at a network edge, the application can be deployed in a container and one or more model versions stored in local storage at the edge, which can be mounted into the container as necessary. When a different model version is to be used, a configuration change or new context can be used to trigger the application to automatically change to the different model version. This updating can be performed seamlessly, without any loss of data.

Abstract: A method, computer program product, and system are provided for associating one or more memory buffers in a computing system with a plurality of memory channels. The method can include associating a first memory buffer to a first plurality of memory banks, where the first plurality of memory banks spans over a first set of one or more memory channels. Similarly, the method can include associating a second memory buffer to a second plurality of memory banks, where the second plurality of memory banks spans over a second set of one or more memory channels. The method can also include associating a first sequence identifier and a second sequence identifier with the first memory buffer and the second memory buffer, respectively. Further, the method can include accessing the first and second memory buffers based on the first and second sequence identifiers.

Abstract: A method for use in a processor for arbitrating between multiple processes to select wavefronts for execution on a shader core is provided. The processor includes a compute pipeline configured to issue wavefronts to the shader core for execution, a hardware queue descriptor associated with the compute pipeline, and the shader core. The shader core is configured to execute work for the compute pipeline corresponding to a first memory queue descriptor executed using data for the first memory queue descriptor that is loaded into a first hardware queue descriptor. The processor is configured to detect a context switch condition, and, responsive to the context switch condition, perform a context switch operation including loading data for a second memory queue descriptor into the first hardware queue descriptor. The shader core is configured to execute work corresponding to the second memory queue descriptor that is loaded into the first hardware queue descriptor.

Abstract: Methods and apparatus are described. A method includes an accelerated processing device running a process. When a maximum time interval during which the process is permitted to run expires before the process completes, the accelerated processing device receives an operating-system-initiated instruction to stop running the process. The accelerated processing device stops the process from running in response to the received operating-system-initiated instruction.

Abstract: A method for use in a processor for arbitrating between multiple processes to select wavefronts for execution on a shader core is provided. The processor includes a compute pipeline configured to issue wavefronts to the shader core for execution, a hardware queue descriptor associated with the compute pipeline, and the shader core. The shader core is configured to execute work for the compute pipeline corresponding to a first memory queue descriptor executed using data for the first memory queue descriptor that is loaded into a first hardware queue descriptor. The processor is configured to detect a context switch condition, and, responsive to the context switch condition, perform a context switch operation including loading data for a second memory queue descriptor into the first hardware queue descriptor. The shader core is configured to execute work corresponding to the second memory queue descriptor that is loaded into the first hardware queue descriptor.

Abstract: A method of processing commands is provided. The method includes holding commands in queues and executing the commands in an order based on their respective priority. Commands having the same priority are held in the same queue.

Abstract: A method, computer program product, and system are provided for associating one or more memory buffers in a computing system with a plurality of memory channels. The method can include associating a first memory buffer to a first plurality of memory banks, where the first plurality of memory banks spans over a first set of one or more memory channels. Similarly, the method can include associating a second memory buffer to a second plurality of memory banks, where the second plurality of memory banks spans over a second set of one or more memory channels. The method can also include associating a first sequence identifier and a second sequence identifier with the first memory buffer and the second memory buffer, respectively. Further, the method can include accessing the first and second memory buffers based on the first and second sequence identifiers.

Abstract: A processor device includes a cache and a memory storing a set of counters. Each counter of the set is associated with a corresponding block of a plurality of blocks of the cache. The processor device further includes a cache access monitor to, for each time quantum for a series of one or more time quanta, increment counter values of the set of counters based on accesses to the corresponding blocks of the cache. The processor device further includes a transfer engine to, after completion of each time quantum, transfer the counter values of the set of counters for the time quantum to a corresponding location in a system memory.

Abstract: A processor device includes a cache and a memory storing a set of counters. Each counter of the set is associated with a corresponding block of a plurality of blocks of the cache. The processor device further includes a cache access monitor to, for each time quantum for a series of one or more time quanta, increment counter values of the set of counters based on accesses to the corresponding blocks of the cache. The processor device further includes a transfer engine to, after completion of each time quantum, transfer the counter values of the set of counters for the time quantum to a corresponding location in a system memory.

Abstract: Methods and apparatus are described. A method includes an accelerated processing device running a process. When a maximum time interval during which the process is permitted to run expires before the process completes, the accelerated processing device receives an operating-system-initiated instruction to stop running the process. The accelerated processing device stops the process from running in response to the received operating-system-initiated instruction.

Abstract: A method of processing commands is provided. The method includes holding commands in queues and executing the commands in an order based on their respective priority. Commands having the same priority are held in the same queue.

Abstract: A processor uses the same virtual address space for heterogeneous processing units of the processor. The processor employs different sets of page tables for different types of processing units, such as a CPU and a GPU, wherein a memory management unit uses each set of page tables to translate virtual addresses of the virtual address space to corresponding physical addresses of memory modules associated with the processor. As data is migrated between memory modules, the physical addresses in the page tables can be updated to reflect the physical location of the data for each processing unit.

Abstract: A method of processing commands is provided. The method includes holding commands in queues and executing the commands in an order based on their respective priority. Commands having the same priority are held in the same queue.

Abstract: A method resumes an accelerated processing device (APD) wavefront in which a subset of elements have faulted. A restore command for a job including a wavefront is received. A list of context states for the wavefront is read from a memory associated with a APD. An empty shell wavefront is created for restoring the list of context states. A portion of not acknowledged data is masked over a portion of acknowledged data within the restored wavefronts.

Abstract: Methods, systems, and computer readable media embodiments are disclosed for preemptive context-switching of processes running on a accelerated processing device. Embodiments include, detecting by an accelerated processing device a memory exception, and preempting a process from running on the accelerated processing device based upon the detected exception.

Abstract: Methods, systems, and computer readable media embodiments are disclosed for preemptive context-switching of processes running on an accelerated processing device. A method includes, responsive to an exception upon access to a memory by a process running on a accelerated processing device, whether to preempt the process based on the exception, and preempting, based upon the determining, the process from running on the accelerated processing device.

Abstract: A method, system, and computer program product are disclosed for providing improved access to accelerated processing device compute resources to user mode applications. The functionality disclosed allows user mode applications to provide commands to an accelerated processing device without the need for kernel mode transitions in order to access a unified ring buffer. Instead, applications are each provided with their own buffers, which the accelerated processing device hardware can access to process commands. With full operating system support, user mode applications are able to utilize the accelerated processing device in much the same way as a CPU.

Abstract: Embodiments describe herein provide a method of for managing task scheduling on a accelerated processing device. The method includes executing a first task within the accelerated processing device (APD), monitoring for an interruption of the execution of the first task, and switching to a second task when an interruption is detected.

Abstract: The desire to use an Accelerated Processing Device (APD) for general computation has increased due to the APD's exemplary performance characteristics. However, current systems incur high overhead when dispatching work to the APD because a process cannot be efficiently identified or preempted. The occupying of the APD by a rogue process for arbitrary amounts of time can prevent the effective utilization of the available system capacity and can reduce the processing progress of the system. Embodiments described herein can overcome this deficiency by enabling the system software to pre-empt a process executing on the APD for any reason. The APD provides an interface for initiating such a pre-emption. This interface exposes an urgency of the request which determines whether the process being preempted is allowed a grace period to complete its issued work before being forced off the hardware.