Google Inc.

Nvidia Corporation

Timothy James Murray

Rudy Jason Sams

Stephen Roderick Hines

Vivek Kini

Forrest Iandola

Joel Adam Scherpelz

Christopher Michael Cameron

One embodiment sets forth a method for assigning priorities to kernels launched by a software application and executed within a stream of work on a parallel processing subsystem. First, the software application assigns a desired priority to a stream using a call included in the API. The API receives this call and passes it to a driver. The driver maps the desired priority to an appropriate device priority associated with the parallel processing subsystem. Subsequently, if the software application launches a particular kernel within the stream, then the driver assigns the device priority associated with the stream to the kernel before adding the kernel to the stream for execution on the parallel processing subsystem. Advantageously, by assigning priorities to streams and, subsequently, strategically launching kernels within the prioritized streams, an application developer may fine-tune the software application to increase the overall processing efficiency of the software application.

Assigning priorities to computational work streams by mapping desired execution priorities to device priorities

A technique for executing a plurality of applications on a GPU. The technique involves establishing a first connection to a first application and a second connection to a second application, establishing a universal processing context that is shared by the first application and the second application, transmitting a first workload pointer to a first queue allocated to the first application, the first workload pointer pointing to a first workload generated by the first application, transmitting a second workload pointer to a second queue allocated to the second application, the second workload pointer pointing to a second workload generated by the second application, transmitting the first workload pointer to a first GPU queue in the GPU, and transmitting the second workload pointer to a second GPU queue in the GPU, wherein the GPU is configured to execute the first workload and the second workload in accordance with the universal processing context.

Graphics processing unit sharing between many applications

An application executing on a computing device may invoke a function call of a first function. The computing device may support a bundled application library of functions and a native library of functions. It may be determined that code for executing the first function exists in both the bundled application library and the native library, and the application may execute the first function using the code in the native library. While the application continues execution on the computing device, the application may invoke a function call of a second function. It may be determined that, of the bundled application library and the native library, the code for executing the second function exists in the bundled application library. The application may execute the second function using the code in the bundled application library.

Application access to native and bundled libraries

Methods and systems for providing compatibility of applications with multiple versions of an operating system are provided. The method may comprise receiving, by a device, a request to perform a function-call of an application that references features of a first version of an operating system. The method also includes determining whether a version of the operating system and a system library on the device supports the function-call of the application so as to be able to perform dynamic compilation of code in the system library for the function-call at run-time, and based on the determination, accessing a compatibility library associated with the application that includes pre-compiled code for execution by a given device having a given version of the operating system other than the first version of the operating system. The method also includes at run-time, performing the function-call using corresponding pre-compiled code from the compatibility library.

Growing community of inventors

San Francisco, CA, United States of America

Timothy James Murray