Improve arp_responder.py performance #17280
Open
+36
−3
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Instead of having scapy.sendp create a socket, bind it, send the packet, and then close the packet every time, just reuse the listening socket that's been created. Signed-off-by: Saikrishna Arcot <[email protected]>
Use the libpcap backend in arp_responder, which sets up a ring buffer for incoming packets. This is expected to be more efficient than making many recvmsg syscalls, especially during ARP/NDP floods. Signed-off-by: Saikrishna Arcot <[email protected]>
Enhance the BPF filter for arp_responder to only get ARP/ICMP packets for the IPv4/IPv6 addresses that it is interested in. This is especially useful in large VLANs with many hosts. Signed-off-by: Saikrishna Arcot <[email protected]>
|
/azp run |
|
Azure Pipelines successfully started running 1 pipeline(s). |
8 tasks
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description of PR
Summary:
Fixes # (issue)
Type of change
Back port request
Approach
What is the motivation for this PR?
Improve the overall performance of arp_responder.py.
How did you do it?
recvmsgsyscalls to get packets one at a time, a RX ring buffer is set up as shared memory between userspace and kernel space. The kernel then writes incoming packets into that ring buffer, and then userspace reads it from there. All this happens with only one syscall toselect(or equivalent) to know when data (one or more packets) is available, instead of 2 syscalls per packet. The downside is that this does increase memory usage, for the ring buffer to be created. A similar thing can be done for TX, but is a bit harder and is currently not supported by libpcap (so packets being sent out will still use the regularsendmsgsyscall.How did you verify/test it?
Tested with multiple advanced reboot warm upgrade runs.
Any platform specific information?
Supported testbed topology if it's a new test case?
Documentation