System and method for directly executing code from block-based memory
AssigneeNVIDIA Corporation
InventorsShang-Tse Chuang, Stephen D. Lew, Gerrit A. Slavenburg
Life of this patent
Estimating the term…
Abstract
A system, method, and computer program product are provided for directly executing code in block-based memory, which resides in communication with a processor and a controller. Utilizing the controller, a request is received from the processor for a subset of a block of data in the block-based memory, and at least a portion of the block is retrieved from the block-based memory. After the retrieval, at least a portion of the block is stored in a cache. The subset of the block is then transmitted to the processor, utilizing the controller. To this end, code in the block-based memory is directly executed.
Claims25
A method, comprising:
- receiving a request from a processor for a subset of a block of data in block-based memory, utilizing a controller in communication with the processor and the block-based memory;
- retrieving at least a portion of the block from the block-based memory, utilizing the controller;
- storing at least a portion of the block in a cache;
- and
- transmitting the subset of the block to the processor, utilizing the controller;
- wherein code in the block-based memory is directly executed by executing application code directly from the block-based memory, while avoiding a requirement that an entirety of an application be copied from the block-based memory prior to the executing thereby avoiding a need of memory on a chip including the processor.
A computer program product embodied on a computer readable medium, comprising:
- computer code for receiving a request from a processor for a subset of a block of data in block-based memory, utilizing a controller in communication with the processor and the block-based memory;
- computer code for retrieving at least a portion of the block from the block-based memory, utilizing the controller;
- computer code for storing at least a portion of the block in a cache;
- and
- computer code for transmitting the subset of the block to the processor, utilizing the controller;
- wherein code in the block-based memory is directly executed by executing application code directly from the block-based memory, while avoiding a requirement that an entirety of an application be copied from the block-based memory prior to the executing, thereby avoiding a need of memory on a chip including the processor.
An apparatus, comprising:
- a controller in communication with a processor and block-based memory, the controller, under the control of the processor without intervention from other memory, capable of receiving a request from the processor for a subset of a block of data in the block-based memory, retrieving at least a portion of the block from the block-based memory, storing at least a portion of the block in a cache, and transmitting the subset of the block to the processor;
- wherein code in the block-based memory is directly executed by executing application code directly from the block-based memory, while avoiding a requirement that an entirety of an application be copied from the block-based memory prior to the executing, thereby avoiding a need of memory on a chip including the processor.
Dependencies are inferred from the claim text.