# US 2019/0317891 A1 · Ordering updates for nonvolatile memory accesses

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

- Number: US 2019/0317891 A1
- Document ID: US20190317891A1
- Status: Published application
- Kind: A1
- Application number: 16/453,784
- Filed: 2019-06-26
- Published: 2019-10-17
- Earliest claimed: 2017-07-24
- Assignees: Hewlett Packard Enterprise Development LP
- Applicants: Hewlett Packard Enterprise Development LP
- Inventors: Sanketh Nalli; Haris Volos; Kimberly Keeton
- Web page: https://patentgrep.com/patent/US20190317891A1

## CPC

- G06F 12/0238: {Memory management in non-volatile memory, e.g. resistive RAM or ferroelectric memory}
- G06F 12/0804: with main memory updating (G06F12/0806 takes precedence)
- G06F 12/0868: Data transfer between cache memory and other subsystems, e.g. storage devices or host systems
- G06F 2212/1028: Power efficiency
- G06F 2212/1032: Reliability improvement, data loss prevention, degraded operation etc
- G06F 2212/7203: Temporary buffering, e.g. using volatile buffer or dedicated buffer blocks
- Y02D 10/00: Energy efficient computing, e.g. low power processors, power management or thermal management
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- Y02D: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE

## Abstract

Examples relate to ordering updates for nonvolatile memory accesses. In some examples, a first update that is propagated from a write-through processor cache of a processor is received by a write ordering buffer, where the first update is associated with a first epoch. The first update is stored in a first buffer entry of the write ordering buffer. At this stage, a second update that is propagated from the write-through processor cache is received, where the second update is associated with a second epoch. A second buffer entry of the write ordering buffer is allocated to store the second update. The first buffer entry and the second buffer entry can then be evicted to non-volatile memory in epoch order.

## Claims

### Claim at position 1

-15. (canceled)

### Claim 16

A system for processing ordered updates in nonvolatile memory access, the system comprising:
a processor;
a processor cache that is operatively connected to the processor;
a non-volatile memory;
a write ordering buffer that is operatively connected to the processor and the non-volatile memory;
wherein the processor is configured to:
receive data for propagation to the processor cache and the write ordering buffer;
determine if each logical portion of data is a cache hit;
in response to determining a portion of data is a write miss, determine if the write ordering buffer has space for a new buffer entry for an epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, create the new buffer entry for the epoch corresponding to the data; and
write a cache update to the new buffer entry.

### Claim 17 (depends on 16)

The system of claim 16, wherein, in response to determining a portion of data is a write miss, the processor is further configured to not create a new cache line entry in the processor cache.

### Claim 18 (depends on 16)

The system of claim 16, wherein the new buffer entry comprises a size equivalent to a size of a cache line of the processor cache.

### Claim 19 (depends on 16)

The system of claim 16, the processor further configured to, in response to determining a portion of data is a cache hit, determine if a buffer entry for an epoch corresponding to the data exists in the write ordering buffer.

### Claim 20 (depends on 19)

The system of claim 19, the processor further configured to:
in response to determining a buffer entry for the epoch corresponding to the data does not exist, determine if the write ordering buffer has space for a new buffer entry for the epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, create the new buffer entry for the epoch corresponding to the data; and
write a cache update to the new buffer entry.

### Claim 21 (depends on 19)

The system of claim 19, the processor further configured to:
in response to determining a buffer entry for the epoch corresponding to the data does exist, determine if the write ordering buffer has space for a new buffer entry for the epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, create the new buffer entry for the epoch corresponding to the data; and
write a cache update to the new buffer entry.

### Claim 22 (depends on 19)

The system of claim 19, each buffer entry of a plurality of buffer entries of the write ordering buffer comprising at least one dirty bit.

### Claim 23 (depends on 22)

The system of claim 22, wherein the dirty bit specifies that a subset of a buffer entry is dirty, and wherein the subset is a byte or word of the buffer entry.

### Claim 24 (depends on 16)

The system of claim 16, wherein the write-through processor cache implements a no-allocate write policy.

### Claim 25

A method for processing ordered updates in nonvolatile memory access, the method comprising:
receiving, by a processor, data for propagation to a processor cache and a write ordering buffer;
determining if each logical portion of data is a cache hit;
in response to determining a portion of data is a write miss, determining if the write ordering buffer has space for a new buffer entry for an epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, creating the new buffer entry for the epoch corresponding to the data; and
writing a cache update to the new buffer entry.

### Claim 26 (depends on 25)

The method of claim 25, wherein, in response to determining a portion of data is a write miss, creating a new cache line entry in the processor cache.

### Claim 27 (depends on 25)

The method of claim 25, wherein the new buffer entry comprises a size equivalent to a size of a cache line of the processor cache.

### Claim 28 (depends on 25)

The method of claim 25, further comprising, in response to determining a portion of data is a cache hit, determining if a buffer entry for an epoch corresponding to the data exists in the write ordering buffer.

### Claim 29 (depends on 28)

The method of claim 28, further comprising:
in response to determining a buffer entry for the epoch corresponding to the data does not exist, determining if the write ordering buffer has space for a new buffer entry for the epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, creating the new buffer entry for the epoch corresponding to the data; and
writing a cache update to the new buffer entry.

### Claim 30 (depends on 28)

The method of claim 28, further comprising:
in response to determining a buffer entry for the epoch corresponding to the data does exist, determining if the write ordering buffer has space for a new buffer entry for the epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, creating the new buffer entry for the epoch corresponding to the data; and
writing a cache update to the new buffer entry.

### Claim 31 (depends on 28)

The method of claim 28, each buffer entry of a plurality of buffer entries of the write ordering buffer comprising at least one dirty bit.

### Claim 32 (depends on 31)

The method of claim 31, wherein the dirty bit specifies that a subset of a buffer entry is dirty, and wherein the subset is a byte or word of the buffer entry.

### Claim 33 (depends on 25)

The method of claim 25, wherein the write-through processor cache implements a no-allocate write policy.

### Claim 34

A non-transitory machine-readable storage medium encoded with instructions executable by a processor for processing ordered updates in nonvolatile memory access, the machine-readable storage medium comprising instructions to:
receive data for propagation to a processor cache and a write ordering buffer;
determine if each logical portion of data is a cache hit;
in response to determining a portion of data is a write miss, determine if the write ordering buffer has space for a new buffer entry for an epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, create the new buffer entry for the epoch corresponding to the data; and
write a cache update to the new buffer entry.

### Claim 35 (depends on 34)

The non-transitory machine-readable storage medium of claim 34, the machine-readable storage medium further comprising instructions to:
in response to determining a portion of data is a cache hit, determine if a buffer entry for an epoch corresponding to the data exists in the write ordering buffer;
in response to determining a buffer entry for the epoch corresponding to the data does not exist, determine if the write ordering buffer has space for a new buffer entry for the epoch corresponding to the data;
in response to determining the write ordering buffer has space for the new buffer entry, create the new buffer entry for the epoch corresponding to the data; and
write a cache update to the new buffer entry.
