# US 2012/0254896 A1 · Method and system for protocol offload in paravirtualized systems

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

- Number: US 2012/0254896 A1
- Document ID: US20120254896A1
- Status: Published application
- Kind: A1
- Application number: 13/523,088
- Filed: 2012-06-14
- Published: 2012-10-04
- Earliest claimed: 2006-11-28
- Assignees: Broadcom Corporation
- Inventors: Eliezer Aloni; Rafi Shalom
- Web page: https://patentgrep.com/patent/US20120254896A1

## CPC

- G06F 9/45533: {Hypervisors; Virtual machine monitors}
- G06F 9/4555: {Para-virtualisation, i.e. guest operating system has to be modified}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Certain aspects of a method and system for protocol offload in paravirtualized systems may be disclosed. Exemplary aspects of the method may include preposting of application buffers to a front-end driver rather than to a NIC in a paravirtualized system. The NIC may be enabled to place the received offloaded data packets into a received data buffer corresponding to a particular guest operating system (GOS). A back-end driver may be enabled to acknowledge the placed offloaded data packets. The back-end driver may be enabled to forward the received data buffer corresponding to the particular GOS to the front-end driver. The front-end driver may be enabled to copy offloaded data packets from a received data buffer corresponding to a particular GOS to the preposted application buffers.

## Claims

### Claim 1

A system for processing data, the system comprising:
a guest operation system (GOS) configured to prepost an application buffer to a front-end driver;
a hypervisor; and
a network interface card (NIC) configured to allocate a memory buffer for the guest operating system, to post the allocated memory buffer for the guest operating system to a receive queue, and to place received offloaded data packets into the memory buffer,
wherein a back-end driver of the hypervisor is enabled to forward the memory buffer to a front-end driver associated with the guest operation system and the front-end driver is enabled to copy the received offloaded data packets from the memory buffer to the application buffer.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the network interface card places the received offloaded data packets into the memory buffer based on a MAC address associated with the received offloaded data packets.

### Claim 3 (depends on 1)

The system according to claim 1, wherein the network interface card places the received offloaded data packets into the memory buffer based on a GOS identification parameter that identifies a particular guest operating system.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the network interface card is operable to place the received offloaded data packets into the memory buffer based on a network connection parameter associated with the received data packets.

### Claim 5 (depends on 1)

The system according to claim 1, wherein the back-end driver is enabled to acknowledge the placement of the received offloaded data packets by the network interface card.

### Claim 6 (depends on 1)

The system according to claim 1, wherein the network interface card places the received offload data packets into the memory buffer utilizing direct memory access.

### Claim 7 (depends on 1)

The system according to claim 1, wherein the preposted application buffer are stored at the front-end driver without posting the preposted application buffer to the back-end driver.

### Claim 8 (depends on 1)

The system according to claim 1, wherein the back-end driver limits a number of offloaded network connections based on a network connection parameter associated with the placed received data packets.

### Claim 9 (depends on 8)

The system according to claim 8, wherein the offloaded network connections is one of a TCP/IP connection, RDMA connection, or an iSCSI connection.

### Claim 10

A method for processing data, the system comprising:
preposting an application buffer from a guest operating system to a front-end driver;
allocating, by a network interface card, a memory buffer for each guest operating system;
posting the allocated memory buffer for the guest operating system to a receive queue;
placing received offloaded data packets into the memory buffer; and
copying, by the front-end driver, the received offloaded data packets from the memory buffer to the application buffer.

### Claim 11 (depends on 10)

The method according to claim 10, wherein the received offloaded data packets are placed into the memory buffer based on a MAC address associated with the received offloaded data packets.

### Claim 12 (depends on 10)

The method according to claim 10, wherein the received offloaded data packets are placed into the memory buffer based on a GOS identification parameter that identifies a particular guest operating system.

### Claim 13 (depends on 10)

The method according to claim 10, wherein the received offloaded data packets are placed into the memory buffer based on a network connection parameter associated with the received data packets.

### Claim 14 (depends on 10)

The method according to claim 10, further comprising acknowledging, via the back-end driver, the placement of the received offloaded data packets by the network interface card.

### Claim 15 (depends on 10)

The method according to claim 10, wherein the received offload data packets are placed into the memory buffer utilizing direct memory access.

### Claim 16 (depends on 10)

The method according to claim 10, further comprising storing the preposted application buffer at the front-end driver without posting the preposted application buffer to the back-end driver.

### Claim 17 (depends on 10)

The method according to claim 10, further comprising limiting a number of offloaded network connections based on a network connection parameter associated with the placed received data packets.

### Claim 18 (depends on 17)

The method according to claim 17, wherein the offloaded network connections is one of a TCP/IP connection, RDMA connection, or an iSCSI connection.

### Claim 19

A method for processing data, the method comprising:
receiving an acknowledgement that received data packets have been placed in one or more memory buffers corresponding to a particular guest operating system (GOS) by a network interface card (NIC); and
responsive to receiving the acknowledgment, copying the one or more memory buffers to a preposted application buffer by a front-end driver.

### Claim 20 (depends on 19)

The method of claim 19, further comprising preposting the application buffer to the front-end driver by the GOS.
