# US 2019/0171575 A1 · Device, system and method to facilitate disaster recovery for a multi-processor platform

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

- Number: US 2019/0171575 A1
- Document ID: US20190171575A1
- Status: Published application
- Kind: A1
- Application number: 16/264,447
- Filed: 2019-01-31
- Published: 2019-06-06
- Earliest claimed: 2019-01-31
- Assignees: Intel Corporation
- Inventors: Wei CHEN; Eswaramoorthi NALLUSAMY; Larisa NOVAKOVSKY; Mark SCHMISSEUR; Eric RASMUSSEN; Stephen VAN DOREN; Yen-Cheng LIU
- Web page: https://patentgrep.com/patent/US20190171575A1

## CPC

- G06F 1/30: Means for acting in the event of power-supply failure or interruption, e.g. power-supply fluctuations (for resetting only G06F1/24)
- G06F 12/0891: using clearing, invalidating or resetting means
- G06F 11/1441: {Resetting or repowering}
- G06F 12/0246: {in block erasable memory, e.g. flash memory}
- G06F 12/0802: Addressing of a memory level in which the access to the desired data or data block requires associative addressing means, e.g. caches
- G06F 12/0897: with two or more cache hierarchy levels (with multilevel cache hierarchies G06F12/0811)
- G06F 2212/7205: Cleaning, compaction, garbage collection, erase control
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Techniques and mechanisms for capturing an image of processor state at one node of multiple nodes of a multi-processor platform, where the processor state includes some version of data which the node retrieved from another node of the platform. In an embodiment, a disruption of power is detected when a processor of a first node has a cached version of data which was retrieved from a second node. In response to detection of the disruption, the data is saved to a system memory of the first node as part of an image of the processor's state. The image further comprises address information, corresponding to the data, which indicates a memory location at the second node. In another embodiment, processor state is restored during a boot-up of the node, wherein the state includes the captured version of data which was previously retrieved from the second node.

## Claims

### Claim 1

A circuit device comprising:
a first processor to couple to a first memory, wherein a first node is to comprise the first processor and the first memory, wherein the first processor comprises:
first circuitry coupled to detect a power disruption event while first data is cached at the first processor and while the first node is coupled to a second node which comprises a second processor and a second memory, wherein the first data corresponds to first address information which indicates a memory location of the second memory; and
second circuitry, responsive to the first circuitry, to:
interrupt a pipeline of the first processor;
terminate communications between the first node and the second node; and
capture an image of a state of the first processor, comprising the second circuitry to flush a cache of the first processor to a reserved region of the first memory, wherein the second circuitry is to write both the first data and the first address information to the reserved region.

### Claim 2 (depends on 1)

The circuit device of claim 1, wherein the first processor further comprises:
third circuitry to generate a signal after the image is captured, the signal to enable a shutdown of the first processor.

### Claim 3 (depends on 1)

The circuit device of claim 1, wherein the first circuitry is to detect the power disruption event while second data is cached at the first processor, wherein the second data corresponds to second address information which represents a memory location of the first memory, wherein the second circuitry to flush the cache further comprises the second circuitry to write both the second data and the second address information to the reserved region.

### Claim 4 (depends on 1)

The circuit device of claim 1, further comprising the second node, wherein the first circuitry is to detect the power disruption event while second data is cached at the second processor, wherein the second data corresponds to second address information which indicates a second memory location of the first memory, wherein the second processor comprises:
third circuitry which, responsive to an indication of the power disruption event, is to:
interrupt a pipeline of the second processor;
capture an image of a state of the second processor, comprising the third circuitry to flush a second cache of the second processor to a second reserved region of the second memory, wherein the third circuitry to flush the second cache comprises the third circuitry to write both the second data and the second address information to the second reserved region.

### Claim 5 (depends on 1)

The circuit device of claim 1, wherein the second circuitry is to capture the image according to an allocation of multiple caches of the first processor each to a different respective sub-region of the reserved region.

### Claim 6 (depends on 1)

The circuit device of claim 1, wherein the second circuitry to terminate the communications comprises the second circuitry to receive an acknowledgement that a link between the first node and the second node is blocked, wherein, responsive to the first circuitry, the second circuitry is further to flush the pipeline of the first processor before the second circuitry is to capture the image.

### Claim 7 (depends on 1)

The circuit device of claim 1, wherein the second circuitry to flush the cache of the first processor to the reserved region comprises the second circuitry to preserve a relative order of data cached by respective lines of the cache.

### Claim 8 (depends on 1)

The circuit device of claim 1, wherein any cached data which is to be flushed from the first processor to the second node is to be flushed independent of the power disruption event.

### Claim 9 (depends on 1)

The circuit device of claim 1, wherein the first processor further comprises:
third circuitry to detect a boot-up of the first processor;
fourth circuitry, responsive to the third circuitry, to:
perform an access of the image at the reserved region; and
based on the access, to write the first data and the first address information to the cache of the first processor.

### Claim 10 (depends on 1)

The circuit device of claim 1, wherein the second circuitry to capture the image comprises the second circuitry to write the image to a volatile memory.

### Claim 11 (depends on 10)

The circuit device of claim 10, wherein the second circuitry to capture the image further comprises the second circuitry to write the image to a non-volatile memory after the image is written to the volatile memory.

### Claim 12 (depends on 1)

The circuit device of claim 1, further comprising:
third circuitry, coupled between a core of the first processor and the first memory, to perform a hash calculation, wherein the second circuitry is further to disable the third circuitry in response to the first circuitry, wherein the second circuitry is to capture the image while the third circuitry is disabled.

### Claim 13

A system comprising:
a first node comprising:
a first memory; and
a first processor couple to the first memory, wherein the first processor comprises:
first circuitry coupled to detect a power disruption event while first data is cached at the first processor and while the first node is coupled to a second node which comprises a second processor and a second memory, wherein the first data corresponds to first address information which indicates a memory location of the second memory; and
second circuitry, responsive to the first circuitry, to:
interrupt a pipeline of the first processor;
terminate communications between the first node and the second node; and
capture an image of a state of the first processor, comprising the second circuitry to flush a cache of the first processor to a reserved region of the first memory, wherein the second circuitry is to write both the first data and the first address information to the reserved region; and
a display device coupled to the first node, the display device to display an image based on the first data.

### Claim 14 (depends on 13)

The system of claim 13, wherein the second circuitry is to capture the image according to an allocation of multiple caches of the first processor each to a different respective sub-region of the reserved region.

### Claim 15 (depends on 13)

The system of claim 13, wherein the second circuitry to flush the cache of the first processor to the reserved region comprises the second circuitry to preserve a relative order of data cached by respective lines of the cache.

### Claim 16 (depends on 13)

The system of claim 13, wherein the first processor further comprises:
third circuitry to detect a boot-up of the first processor;
fourth circuitry, responsive to the third circuitry, to:
perform an access of the image at the reserved region; and
based on the access, to write the first data and the first address information to the cache of the first processor.

### Claim 17 (depends on 13)

The system of claim 13, wherein the first node further comprises:
third circuitry, coupled between a core of the first processor and the first memory, to perform a hash calculation, wherein the second circuitry is further to disable the third circuitry in response to the first circuitry, wherein the second circuitry is to capture the image while the third circuitry is disabled.

### Claim 18

A method at a circuit device, the method comprising:
detecting a power disruption event, the detecting while first data is cached at a first processor of a first node, wherein a first memory of the first node is accessible to the first processor, wherein the first node is coupled to a second node which comprises a second processor and a second memory, wherein the first data corresponds to first address information which indicates a memory location of the second memory;
based on the detecting of the power disruption event:
interrupting a pipeline of the first processor;
terminating communications between the first node and the second node; and
capturing an image of a state of the first processor, comprising flushing a cache of the first processor to a reserved region of the first memory, wherein the flushing the cache comprises writing both the first data and the first address information to the reserved region.

### Claim 19 (depends on 18)

The method of claim 18, wherein the image is captured according to an allocation of multiple caches of the first processor each to a different respective sub-region of the reserved region.

### Claim 20 (depends on 18)

The method of claim 18, wherein flushing the cache of the first processor to the reserved region preserves a relative order of data cached by respective lines of the cache.

### Claim 21 (depends on 18)

The method of claim 18, wherein any cached data which is flushed from the first processor to the second node is flushed independent of the detecting of the power disruption event.

### Claim 22 (depends on 18)

The method of claim 18, further comprising:
detecting a boot-up of the first processor;
based on the detecting of the boot-up, restoring the state of the processor, including:
accessing the image at the reserved region; and
based on the accessing, writing the first data and the first address information to the cache of the first processor.

### Claim 23 (depends on 18)

The method of claim 18, further comprising:
based on the detecting of the power disruption event, disabling a hash engine which is coupled between a core of the first processor and the first memory, wherein the image is captured while the hash engine is disabled.
