Qualcomm Incorporated

Colin Christopher Sharp

David Rigel Garcia Garcia

Andrew Evan Gruber

Alexei Vladimirovich Bourd

Chihong Zhang

Eduardus Antonius Metz

Thomas Andrew Sartorius

Avinash Seetharamaiah

Christopher Paul Frascati

Jonnala Gadda Nagendra Kumar

Murat Balci

Thomas Edwin Frisinger

Eric Demers

Sudeep Ravi Kottilingal

Maurice Franklin Ribble

Ramesh Viswanathan

Bohuslav Rychlik

Moinul H Khan

Chun Yu

Srihari Babu Alla

David A Gotwalt

Nigel Terence Poole

Robert J Simpson

Tzung Ren Tzeng

Jeffrey Leger

Tarun Reddy Karra

Zachary Aaron Pfeffer

Aleksandra L Krstic

Jonathan Gerald Thomason

Maurice Ribble

Eduardus A Metz

A device comprising a graphics processing unit (GPU) includes a memory and at least one processor. The at least one processor may be configured to: receive a GPU command packet that indicates the GPU may select between a direct rendering mode or a binning rendering mode for a portion of a frame to be rendered by the GPU, determine whether to use the direct rendering mode or the binning rendering mode for the portion of the frame to be rendered by the GPU based on at least one of: information in the received command packet or a state of the GPU, and render the portion of the frame using the determined direct rendering mode or the binning rendering mode.

Flex rendering based on a render target in graphics processing

This disclosure proposes techniques for graphics processing. In one example, a graphics processing unit (GPU) is configured to access a memory according to one of an unsecure mode and a secure mode. The GPU may include a memory access controller configured to direct memory transactions from at least one hardware unit of the GPU to an unsecure memory unit or a secure memory unit based on the unsecure mode or secure mode and a resource descriptor associated with a memory resource.

Hardware enforced content protection for graphics processing units

This disclosure is directed to deferred preemption techniques for scheduling graphics processing unit (GPU) command streams for execution on a GPU. A host CPU is described that is configured to control a GPU to perform deferred-preemption scheduling. For example, a host CPU may select one or more locations in a GPU command stream as being one or more locations at which preemption is allowed to occur in response to receiving a preemption notification, and may place one or more tokens in the GPU command stream based on the selected one or more locations. The tokens may indicate to the GPU that preemption is allowed to occur at the selected one or more locations. This disclosure further describes a GPU configured to preempt execution of a GPU command stream based on one or more tokens placed in a GPU command stream.

Deferred preemption techniques for scheduling graphics processing unit command streams

This disclosure describes techniques for extending the architecture of a general purpose graphics processing unit (GPGPU) with parallel processing units to allow efficient processing of pipeline-based applications. The techniques include configuring local memory buffers connected to parallel processing units operating as stages of a processing pipeline to hold data for transfer between the parallel processing units. The local memory buffers allow on-chip, low-power, direct data transfer between the parallel processing units. The local memory buffers may include hardware-based data flow control mechanisms to enable transfer of data between the parallel processing units. In this way, data may be passed directly from one parallel processing unit to the next parallel processing unit in the processing pipeline via the local memory buffers, in effect transforming the parallel processing units into a series of pipeline stages.

Computational resource pipelining in general purpose graphics processing unit

A transparent format converter (TFC) may determine that a request by at least one processor for graphics data stored in graphics memory is indicative of a request for graphics data in a first data format. The TFC may retrieve the graphics data in a second data format from the graphics memory based at least in part on the request for the graphics data in the graphics memory. The TFC may convert the retrieved graphics data from the second data format to the first data format. The TFC may store the converted graphics data in the first data format into a memory that is accessible by the at least one processor.

Transparent pixel format converter

This disclosure proposes techniques for graphics processing. In one example, a graphics processing unit (GPU) is configured to access a memory according to one of an unsecure mode and a secure mode. The GPU may include a memory access controller configured to direct memory transactions from at least one hardware unit of the GPU to a secure context bank in a memory controller when the GPU is operating in a secure mode, and configured to direct memory transactions from the at least one hardware unit of the GPU to an unsecure context bank in the memory controller when the GPU is operating in the unsecure mode.

This disclosure describes communication techniques that may be used within a multiple-processor computing platform. The techniques may, in some examples, provide software interfaces that may be used to support message passing within a multiple-processor computing platform that initiates tasks using command queues. The techniques may, in additional examples, provide software interfaces that may be used for shared memory inter-processor communication within a multiple-processor computing platform. In further examples, the techniques may provide a graphics processing unit (GPU) that includes hardware for supporting message passing and/or shared memory communication between the GPU and a host CPU.

Inter-processor communication techniques in a multiple-processor computing platform

The present disclosure provides for systems and methods to process a non-resident page that may include attempting to access the non-resident page, an address for the non-resident page pointing to a memory page containing default values, determining that the non-resident page should not cause a page fault based on an indicator indicating that a particular non-resident page should not generate a page fault, returning an indication that a memory read did not translate and returning the default value when the access of the non-resident page is a read and the non-resident page should not cause a page fault. Another example may discontinue a write when the access of the non-resident page is a write and the non-resident page should not cause a page fault.

Conditional page fault control for page residency

A device includes a memory that stores a first page table that includes a first page table entry, wherein the first page table entry further includes a physical address, an alternative location associated with the page table entry, and a physical page of memory associated with the physical address. A first processing unit is configured to: read the first page table entry, and determine the physical address from the first page table entry. The second processing unit is configured to: read the physical address from the first page table entry, determine second page attribute data from the alternative location, wherein the second page attribute data define one or more accessibility attributes of the physical page of memory for the second processing unit, and access the physical page of memory associated with the physical address according to the one or more accessibility attributes.

Multi-core page table sets of attribute fields

The techniques described in the disclosure are generally related to gradual, iterative hang recovery for a graphics processing unit (GPU). The techniques described in the disclosure attempt to re-execute instructions of an application in response to a GPU hang, rather than stopping the execution of the application. If the re-execution causes the GPU to hang again, the techniques described in the disclosure cause the GPU to iteratively execute next set of instructions.

Device hang detection and recovery

A method includes storing, with a first programmable processor, shared variable data to cache lines of a first cache of the first processor. The method further includes executing, with the first programmable processor, a store-with-release operation, executing, with a second programmable processor, a load-with-acquire operation, and loading, with the second programmable processor, the value of the shared variable data from a cache of the second programmable processor.

Multi-core compute cache coherency with a release consistency memory ordering model

This disclosure proposes techniques for demand paging for an IO device (e.g., a GPU) that utilize pre-fetch and pre-back notification event signaling to reduce latency associated with demand paging. Page faults are limited by performing the demand paging operations prior to the IO device actually requesting unbacked memory.

GPU memory buffer pre-fetch and pre-back signaling to avoid page-fault

In general, aspects of this disclosure describe example techniques for efficient storage of data of various data types for graphics processing. In some examples, a processing unit may assign first and second contiguous range of addresses for a first and second data type, respectively. The processing unit may store at least one of graphics data of the first or second data type or addresses of the graphics data of the first or second data type within blocks whose addresses are within the first and second contiguous range of addresses, respectively. The processing unit may store, in cache lines of a cache, the graphics data of the first data type, and the graphics data of the second data type.

Growing community of inventors

San Diego, CA, United States of America

Colin Christopher Sharp