# US 2012/0254497 A1 · Method and apparatus to facilitate shared pointers in a heterogeneous platform

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2012/0254497 A1
- Document ID: US20120254497A1
- Status: Published application
- Kind: A1
- Application number: 13/074,779
- Filed: 2011-03-29
- Published: 2012-10-04
- Earliest claimed: 2011-03-29
- Assignees: Intel Corporation
- Inventors: YANG NI; Rajkishore Barik; Ali-Reza Adl-Tabatabai; Tatiana Shpeisman; Jayanth N. Rao; Ben J. Ashbaugh; Tomasz Janczak
- Web page: https://patentgrep.com/patent/US20120254497A1

## CPC

- G06F 12/0806: Multiuser, multiprocessor or multiprocessing cache systems
- G06F 15/167: using a common memory, e.g. mailbox
- G06T 1/60: Memory management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method and apparatus to facilitate shared pointers in a heterogeneous platform. In one embodiment of the invention, the heterogeneous or non-homogeneous platform includes, but is not limited to, a central processing core or unit, a graphics processing core or unit, a digital signal processor, an interface module, and any other form of processing cores. The heterogeneous platform has logic to facilitate sharing of pointers to a location of a memory shared by the CPU and the GPU. By sharing pointers in the heterogeneous platform, the data or information sharing between different cores in the heterogeneous platform can be simplified.

## Claims

### Claim 1

A method comprising:
sharing a pointer to a memory location between a first processing core and a second processing core, wherein the memory location resides in a shared memory region accessible by both the first and the second processing core.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
defining the pointer to the memory location by the first processing core;
determining a first integer constant comprising a base memory address of the memory location by the first processing core; and
determining a second integer parameter comprising a value of a memory address of the pointer to the memory location by the first processing core.

### Claim 3 (depends on 1)

The method of claim 1, wherein sharing the pointer to the memory location between the first processing core and the second processing core comprises:
defining a second pointer based at least in part on the first integer constant and the second integer parameter by the second processing core, wherein the pointer to the memory location and the second pointer point to a same physical memory location.

### Claim 4 (depends on 3)

The method of claim 3, wherein defining a second pointer based at least in part on the first integer constant and the second integer parameter by the second processing core comprises:
defining a third pointer based at least in part on another base memory address of a virtual memory location of the second processing core and the first integer constant; and
defining the second pointer based at least in part on the third pointer and the second integer parameter.

### Claim 5 (depends on 1)

The method of claim 1, wherein the first processing core is a central processing unit (CPU) and the second processing core is a graphics processing unit (GPU).

### Claim 6 (depends on 4)

The method of claim 4, wherein the second pointer and the third pointer of the second processing core are associated with a binding table.

### Claim 7 (depends on 1)

The method of claim 1, wherein the first and second processing cores are compliant at least in part with one of Open Computing Language, low level virtual machine (LLVM) intermediate language (IL), and GenX JIT IL.

### Claim 8

An apparatus comprising:
a plurality of non-homogeneous processing cores;
a memory coupled with each of the plurality of non-homogeneous processing cores; and
logic to facilitate sharing of a pointer to a location of a shared region in the memory by the plurality of non-homogeneous processing cores.

### Claim 9 (depends on 8)

The apparatus of claim 8, wherein a first processing core of the plurality of non-homogeneous processing cores has a first virtual map associated with the shared region, and wherein a second processing core of the plurality of non-homogeneous processing cores has a second virtual map associated with the shared region.

### Claim 10 (depends on 9)

The apparatus of claim 9, wherein the logic is further to:
determine a first pointer to the first virtual map;
determine a first offset comprising a base memory address of the first virtual map; and
determine a second offset comprising a value of a memory address of the first pointer to the first virtual map.

### Claim 11 (depends on 8)

The apparatus of claim 8, wherein the logic to facilitate sharing of the pointer to the location of the shared region in the memory by the plurality of non-homogeneous processing cores is to:
define a second pointer to the second virtual map based at least in part on the first and second offset, wherein the first pointer and the second pointer reference a same physical memory location in the shared region.

### Claim 12 (depends on 11)

The apparatus of claim 11, wherein the logic to define the second pointer to the second virtual map based at least in part on the first and second offset is to:
define a third pointer based at least in part on another base memory address of the second virtual map and the first offset; and
define the second pointer based at least in part on the third pointer and the second offset.

### Claim 13 (depends on 11)

The apparatus of claim 11, wherein the plurality of non-homogeneous processing cores comprises at least one central processing unit (CPU) and at least one graphics processing unit (GPU).

### Claim 14 (depends on 12)

The apparatus of claim 12, wherein the second pointer and the third pointer are associated with a binding table.

### Claim 15 (depends on 8)

The apparatus of claim 8, wherein the plurality of non-homogeneous processing cores are compliant at least in part with one of Open Computing Language, low level virtual machine (LLVM) intermediate language (IL), and GenX JIT IL.

### Claim 16

A machine readable storage medium having instructions stored thereon which, when executed, cause a processor to perform the following method:
sharing a pointer to a memory location between a first processing core and a second processing core, wherein the memory location resides in a shared memory region accessible by both the first and the second processing core.

### Claim 17 (depends on 16)

The medium of claim 16, further comprising:
defining the pointer to the memory location by the first processing core;
determining a first offset comprising a base memory address of the memory location by the first processing core; and
determining a second offset comprising a value of a memory address of the pointer to the memory location by the first processing core.

### Claim 18 (depends on 16)

The medium of claim 16, wherein sharing the pointer to the memory location between the first processing core and the second processing core comprises:
defining a second pointer based at least in part on the first and second offsets by the second processing core, wherein the pointer to the memory location and the second pointer point to a same physical memory location.

### Claim 19 (depends on 18)

The medium of claim 18, wherein defining a second pointer based at least in part on the first and second offsets by the second processing core comprises:
defining a third pointer based at least in part on another base memory address of a virtual memory location of the second processing core and the first offset; and
defining the second pointer based at least in part on the third pointer and the second offset.

### Claim 20 (depends on 16)

The medium of claim 16, wherein the first processing core is a central processing unit (CPU) and the second processing core is a graphics processing unit (GPU).

### Claim 21 (depends on 19)

The medium of claim 19, wherein the second pointer and the third pointer of the second processing core are associated with a binding table.

### Claim 22 (depends on 16)

The medium of claim 16, wherein the first and second processing cores are compliant at least in part with one of Open Computing Language, low level virtual machine (LLVM) intermediate language (IL), and GenX JIT IL.

### Claim 23 (depends on 16)

The medium of claim 16, wherein the pointer is stored in a format suitable for the second processing core.
