Zero-copy packet transmission between virtualized computing instances

1. A method for a host to perform zero-copy packet transmission, wherein the host supports a forwarding element, a first virtualized computing instance and a second virtualized computing instance, and the method comprises:

allocating a transmit (TX) packet buffer to the first virtualized computing instance and a receive (RX) packet buffer to the second virtualized computing instance, wherein the TX packet buffer and the RX packet buffer are both mapped to a shared memory location associated with the host;

accessing, by the first virtualized computing instance, the TX packet buffer to store a packet in the shared memory location, the packet being addressed from the first virtualized computing instance to the second virtualized computing instance;

processing, by the forwarding element connecting the first virtualized computing instance and the second virtualized computing instance, the packet in the shared memory location; and

accessing, by the second virtualized computing instance, the RX packet buffer to receive the packet stored in the shared memory location.

2. The method of claim 1 , wherein accessing the TX packet buffer comprises:

selecting, from multiple TX packet buffers allocated to the first virtualized computing instance, the TX packet buffer based on destination address information associated with the second virtualized computing instance.

3. The method of claim 2 , wherein accessing the TX packet buffer comprises:

accessing a buffer map storing an association between the TX packet buffer and the destination address information, the packet buffer map being maintained by a network interface controller (NIC) driver supported by a guest operating system associated with the first virtualized computing instance.

4. The method of claim 1 , wherein allocating the TX packet buffer and RX packet buffer comprises:

determining that buffer allocation is required based on one or more of the following: (a) detecting a prior packet that is sent from the first virtualized computing instance to the second virtualized computing instance and (b) detecting that a packet-related metric exceeds a predetermined threshold.

5. The method of claim 1 , wherein processing by the forwarding element comprises:

processing, by the forwarding element, the packet in response to detecting a first notification from the first virtualized computing instance; and

sending a second notification to the second virtualized computing instance to cause the second virtualized computing instance to access the RX packet buffer.

6. The method of claim 5 , wherein processing by the forwarding element comprises:

generating digest information associated with the packet prior to sending the second notification; and

based on the digest information, performing integrity verification to determine whether the packet or the first virtualized computing instance is malicious.

7. The method of claim 1 , wherein processing by the forwarding element comprises:

detecting a third notification indicating that the packet has been received or processed by the second virtualized computing instance; and

generating and sending a fourth notification to cause the first virtualized computing instance to release buffer space storing the packet in the TX packet buffer.

8. A non-transitory computer-readable storage medium that includes a set of instructions which, in response to execution by a processor of a host, cause the processor to perform a method of zero-copy packet transmission, wherein the host supports a forwarding element, a first virtualized computing instance and a second virtualized computing instance, and the method comprises:

allocating a transmit (TX) packet buffer to the first virtualized computing instance and a receive (RX) packet buffer to the second virtualized computing instance, wherein the TX packet buffer and the RX packet buffer are both mapped to a shared memory location associated with the host;

accessing, by the first virtualized computing instance, the TX packet buffer to store a packet in the shared memory location, the packet being addressed from the first virtualized computing instance to the second virtualized computing instance;

processing, by the forwarding element connecting the first virtualized computing instance and the second virtualized computing instance, the packet in the shared memory location; and

accessing, by the second virtualized computing instance, the RX packet buffer to receive the packet stored in the shared memory location.

9. The non-transitory computer-readable storage medium of claim 8 , wherein accessing the TX packet buffer comprises:

selecting, from multiple TX packet buffers allocated to the first virtualized computing instance, the TX packet buffer based on destination address information associated with the second virtualized computing instance.

10. The non-transitory computer-readable storage medium of claim 9 , wherein accessing the TX packet buffer comprises:

accessing a buffer map storing an association between the TX packet buffer and the destination address information, the packet buffer map being maintained by a network interface controller (NIC) driver supported by a guest operating system associated with the first virtualized computing instance.

11. The non-transitory computer-readable storage medium of claim 8 , wherein allocating the TX packet buffer and RX packet buffer comprises:

determining that buffer allocation is required based on one or more of the following: (a) detecting a prior packet that is sent from the first virtualized computing instance to the second virtualized computing instance and (b) detecting that a packet-related metric exceeds a predetermined threshold.

12. The non-transitory computer-readable storage medium of claim 8 , wherein processing by the forwarding element comprises:

processing, by the forwarding element, the packet in response to detecting a first notification from the first virtualized computing instance; and

sending a second notification to the second virtualized computing instance to cause the second virtualized computing instance to access the RX packet buffer.

13. The non-transitory computer-readable storage medium of claim 12 , wherein processing by the forwarding element comprises:

generating digest information associated with the packet prior to sending the second notification; and

based on the digest information, performing integrity verification to determine whether the packet or the first virtualized computing instance is malicious.

14. The non-transitory computer-readable storage medium of claim 8 , wherein processing by the forwarding element comprises:

detecting a third notification indicating that the packet has been received or processed by the second virtualized computing instance; and

generating and sending a fourth notification to cause the first virtualized computing instance to release buffer space storing the packet in the TX packet buffer.

15. A computer system, comprising:

a processor; and

a non-transitory computer-readable medium having stored thereon instructions that, when executed by the processor, cause the processor to:

allocate a transmit (TX) packet buffer to a first virtualized computing instance and a receive (RX) packet buffer to a second virtualized computing instance, wherein the TX packet buffer and the RX packet buffer are both mapped to a shared memory location associated with the computer system;

access, by the first virtualized computing instance, the TX packet buffer to store a packet in the shared memory location, the packet being addressed from the first virtualized computing instance to the second virtualized computing instance;

process, by a forwarding element connecting the first virtualized computing instance and the second virtualized computing instance, the packet in the shared memory location; and

access, by the second virtualized computing instance, the RX packet buffer to receive the packet stored in the shared memory location.

16. The computer system of claim 15 , wherein the instructions for accessing the TX packet buffer cause the processor to:

select, from multiple TX packet buffers allocated to the first virtualized computing instance, the TX packet buffer based on destination address information associated with the second virtualized computing instance.

17. The computer system of claim 16 , wherein the instructions for accessing the TX packet buffer cause the processor to:

access a buffer map storing an association between the TX packet buffer and the destination address information, the packet buffer map being maintained by a network interface controller (NIC) driver supported by a guest operating system associated with the first virtualized computing instance.

18. The computer system of claim 15 , wherein the instructions for allocating the TX packet buffer and RX packet buffer cause the processor to:

determine that buffer allocation is required based on one or more of the following: (a) detecting a prior packet that is sent from the first virtualized computing instance to the second virtualized computing instance and (b) detecting that a packet-related metric exceeds a predetermined threshold.

19. The computer system of claim 15 , wherein the instructions for processing by the forwarding element cause the processor to:

process, by the forwarding element, the packet in response to detecting a first notification from the first virtualized computing instance; and

send a second notification to the second virtualized computing instance to cause the second virtualized computing instance to access the RX packet buffer.

20. The computer system of claim 19 , wherein the instructions for processing by the forwarding element cause the processor to:

generate digest information associated with the packet prior to sending the second notification; and

based on the digest information, performing integrity verification to determine whether the packet or the first virtualized computing instance is malicious.

21. The computer system of claim 15 , wherein the instructions for processing by the forwarding element cause the processor to:

detect a third notification indicating that the packet has been received or processed by the second virtualized computing instance; and

generate and send a fourth notification to cause the first virtualized computing instance to release buffer space storing the packet in the TX packet buffer.