# US 2009/0276607 A1 · Virtualization platform with dedicated cache access

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

- Number: US 2009/0276607 A1
- Document ID: US20090276607A1
- Status: Published application
- Kind: A1
- Application number: 12/113,076
- Filed: 2008-04-30
- Published: 2009-11-05
- Earliest claimed: 2008-04-30
- Assignees: Hewlett-Packard Development Company, L.P.
- Inventors: Thomas J. Bonola
- Web page: https://patentgrep.com/patent/US20090276607A1

## CPC

- G06F 9/30047: {Prefetch instructions; cache control instructions}
- G06F 9/3802: {Instruction prefetching}
- G06F 9/383: {Operand prefetching (cache prefetching G06F12/0862)}
- G06F 9/45558: {Hypervisor-specific management and integration aspects}
- G06F 12/126: with special data handling, e.g. priority of data or instructions, handling errors or pinning
- G06F 2009/45583: {Memory management, e.g. access or allocation}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A computing system supports a virtualization platform with dedicated cache access. The computing system is configured for usage with a memory and a cache and comprises an instruction decoder configured to decode a cache-line allocation instruction and control logic. The control logic is coupled to the instruction decoder and controls the computing system to execute a cache-line allocation instruction that loads portions of data and code regions of the memory into dedicated cache-lines of the cache which are exempted from eviction according to a cache controller replacement policy.

## Claims

### Claim 1

A computing system configured for usage with a memory and a cache comprising:
an instruction decoder configured to decode a cache-line allocation instruction; and
a control logic coupled to the instruction decoder that controls the computing system to execute the cache-line allocation instruction comprising loading portions of data and code regions of the memory into dedicated cache-lines of the cache that are exempted from eviction according to a cache controller replacement policy.

### Claim 2 (depends on 1)

The system according to claim 1 further comprising:
the instruction decoder configured to decode the cache-line allocation instruction and a cache-line de-allocation instruction;
the control logic coupled to the instruction decoder that executes the cache-line allocation instruction by attaching a specified region of memory to the dedicated cache-lines, and that executes the cache-line de-allocation instruction by de-attaching the attached specified region of memory whereby participation in the cache controller replacement policy is restored.

### Claim 3 (depends on 2)

The system according to claim 2 further comprising:
an address register coupled to the instruction decoder that specifies a logical address of a target memory region for attaching to the cache for the cache-line allocation instruction and that specifies a logical address of a target memory region for de-attaching to the cache for the cache-line de-allocation instruction;
a count register coupled to the instruction decoder that specifies a range in bytes to attach to the cache for the cache-line allocation instruction and that specifies a range in bytes to de-attach from the cache for the cache-line de-allocation instruction; and
a cache-type register coupled to the instruction decoder that distinguishes between an instruction-cache and a data-cache.

### Claim 4 (depends on 3)

The system according to claim 3 further comprising:
the cache-type register coupled to the instruction decoder that distinguishes between an instruction-cache, a data-cache, and a translation lookaside buffer.

### Claim 5 (depends on 1)

The system according to claim 1 further comprising:
a virtualization platform configured to run a plurality of operating systems simultaneously on at least one physical server; and
a virtualization logic that loads portions of virtualization platform data and code memory regions into dedicated cache-lines that are exempted from eviction according to the cache controller replacement policy.

### Claim 6 (depends on 5)

The system according to claim 5 further comprising:
the virtualization platform configured to perform global resource scheduling between a plurality of guest operating systems comprising allocating cache-lines to data and code regions of the memory whereby predetermined critical operations are executed a predetermined maximum processor core frequency.

### Claim 7 (depends on 6)

The system according to claim 6 further comprising:
the predetermined critical operations are selected from a group consisting of host resource management including memory and processor management, page shadowing, page table management, exception handling, load balancing including runtime load balancing, event handling, and critical resource management.

### Claim 8 (depends on 1)

The system according to claim 1 further comprising:
a physical server communicatively coupled to the virtualization platform and configured to run the operating system plurality.

### Claim 9 (depends on 1)

The system according to claim 1 further comprising:
a network;
the plurality of physical servers communicatively coupled to the network and the virtualization platform, and configured to run the operating system plurality; and
the plurality of operating systems simultaneously running on at least one physical server.

### Claim 10

A computer-executed method for executing an instruction on a computing system comprising:
decoding a cache-line allocation instruction;
executing the cache-line allocation instruction comprising:
loading portions of data and code regions of the memory into dedicated cache-lines of the cache; and
exempting the dedicated cache-lines from eviction according to a cache controller replacement policy.

### Claim 11 (depends on 10)

The method according to claim 10 further comprising:
decoding the cache-line allocation instruction;
executing the cache-line allocation instruction comprising attaching a specified region of memory to the dedicated cache-lines;
decoding a cache-line de-allocation instruction; and
executing the cache-line de-allocation instruction comprising de-attaching the attached specified region of memory whereby participation in the cache controller replacement policy is restored.

### Claim 12 (depends on 11)

The method according to claim 11 further comprising:
executing the cache-line allocation instruction comprising:
specifying a logical address of a target memory region for attaching to the cache;
specifying a range in bytes to attach to the cache; and
distinguishing between an instruction-cache and a data-cache; and
executing the cache-line de-allocation instruction comprising:
specifying a logical address of a target memory region for de-attaching to the cache;
specifying a range in bytes to de-attach from the cache; and
distinguishing between the instruction-cache and the data-cache.

### Claim 13 (depends on 10)

The method according to claim 10 further comprising:
running a plurality of operating systems simultaneously on at least one physical server;
loading portions of virtualized data and code memory regions into dedicated cache-lines; and
exempting the dedicated cache-lines from eviction according to the cache controller replacement policy.

### Claim 14 (depends on 13)

The method according to claim 13 further comprising:
performing global resource scheduling between a plurality of guest operating systems comprising:
allocating cache-lines to data and code regions of the memory; and
executing predetermined critical operations at a predetermined maximum processor core frequency.

### Claim 15 (depends on 14)

The method according to claim 14 further comprising:
executing predetermined critical operations selected from a group consisting of host resource management including memory and processor management, page shadowing, page table management, exception handling, load balancing including runtime load balancing, event handling, and critical resource management.

### Claim 16

A processor configured for usage with a memory and a cache comprising:
an address register that specifies a logical address of a target memory region for attaching to the cache for a cache-line allocation instruction;
a count register that specifies a range in bytes to attach to the cache for the cache-line allocation instruction; and
a control logic coupled to the address register and the count register that controls the processor to execute the cache-line allocation instruction comprising loading portions of data and code regions of the memory into dedicated cache-lines of the cache that are exempted from eviction according to a cache controller replacement policy.

### Claim 17 (depends on 16)

The processor according to claim 16 further comprising:
the address register that specifies a logical address of a target memory region for de-attaching to the cache for a cache-line de-allocation instruction;
the count register that specifies a range in bytes to de-attach from the cache for the cache-line de-allocation instruction; and
the control logic coupled to the address register and the count register that controls the processor to execute the cache-line de-allocation instruction by de-attaching the attached specified region of memory whereby participation in the cache controller replacement policy is restored.

### Claim 18 (depends on 17)

The processor according to claim 17 further comprising:
a cache-type register that distinguishes cache type for the cache allocation instruction and the cache de-allocation instruction.

### Claim 19 (depends on 16)

The processor according to claim 16 further comprising:
a virtualization platform configured to run a plurality of operating systems simultaneously on at least one physical server; and
a virtualization logic that loads portions of virtualization platform data and code memory regions into dedicated cache-lines that are exempted from eviction according to the cache controller replacement policy.

### Claim 20

An article of manufacture comprising:
a controller-usable medium having a computer readable program code embodied in a controller for executing an instruction, the computer readable program code further comprising:
code causing the controller to decode a cache-line allocation instruction;
code causing the controller to execute the cache-line allocation instruction by loading portions of data and code regions of the memory into dedicated cache-lines of the cache, and exempting the dedicated cache-lines from eviction according to a cache controller replacement policy;
code causing the controller to decode a cache-line de-allocation instruction; and
code causing the controller to execute the cache-line de-allocation instruction comprising de-attaching the attached specified region of memory whereby participation in the cache controller replacement policy is restored.
