Storing kernel attributes
AssigneeNVIDIA Corporation
InventorsDavid Anthony Fontaine, Harold Carter Edwards, David Dastous St Hilaire, Ze Long
Life of this patent
Estimating the term…
Abstract
Apparatuses, systems, and techniques to cause one or more software kernels to be performed based, at least in part, on a data structure to indicate one or more attributes of the one or more software kernels. In at least one embodiment, one or more software kernels are caused to be performed based, at least in part, on a data structure to indicate one or more attributes of the one or more software kernels based on, for example, CUDA or other parallel computing platform code.
Claims24
One or more processors, comprising:
- circuitry to, in response to an application programming interface (API) call of at least one of a runtime library or a device driver of one or more graphics processing units (GPUs):
- receive a data structure indicating a variable number of kernel attributes;
- encode one or more identifiers and corresponding values for one or more of the variable number of kernel attributes;
- and
- use the encoded identifiers and corresponding values to cause the one or more GPUs to launch a software kernel.
A system, comprising:
- one or more computers having one or more processors to, in response to an application programming interface (API) call of at least one of a runtime library or a device driver of one or more graphics processing units (GPUs):
- receive a data structure indicating a variable number of kernel attributes;
- encode one or more identifiers and corresponding values for one or more of the variable number of kernel attributes;
- and
- use the encoded identifiers and corresponding values to cause the one or more GPUs to launch a software kernel.
A non-transitory machine readable medium having stored thereon a set of instructions, which if performed by one or more processors, cause the one or more processors to at least, in response to a call to an application programming interface (API) of at least one of a runtime library or a device driver of one or more graphics processing units (GPUs):
- receive a data structure indicating a variable number of kernel attributes;
- encode one or more identifiers and corresponding values for one or more of the variable number of kernel attributes;
- and
- use the encoded identifiers and corresponding values to cause the one or more GPUs to launch a software kernel.
A method, comprising:
- receiving, by at least a runtime library or device driver of one or more graphics processing units (GPUs), an application programming interface (API) call whose parameters comprise a data structure indicating a variable number of kernel attributes;
- causing, by at least the runtime library or device driver, the one or more GPUs to;
- encode one or more identifiers and corresponding values for one or more of the variable number of kernel attributes;
- and
- use the encoded identifiers and corresponding values to cause the one or more GPUs to launch a software kernel.
Dependencies are inferred from the claim text.