Conversation
This is due to hardware offloads - packets do not have correct checksums with the expectation the hardware will fill it in. (Note this problem is already solved within meshnet; following that example.) 1. Disable Hardware Offloads via ethtool: When NewSocketHandler opens an interface, it uses github.com/safchain/ethtool (matching meshnet's approach) to disable active TX checksum and segmentation offloads as well as RX GRO/LRO/checksum offloads so the Linux kernel veth driver does not coalesce incoming segments or leave local frames with partial checksums. 2. Finalize Partial Checksums via PACKET_AUXDATA: Enabled PACKET_AUXDATA on the raw AF_PACKET socket and updated ReadPacket to inspect TP_STATUS_CSUMNOTREADY (skb->ip_summed == CHECKSUM_PARTIAL). If a frame arrives from a peer veth in another container namespace where TX checksum offload was still enabled, finalizePartialChecksum computes the missing L4 (TCP/UDP/ICMPv6) checksum in software before forwarding. 3. Unit Tests: Added unit tests covering offload disabling, PACKET_AUXDATA parsing, and IPv4/IPv6/VLAN partial checksum finalization.
- don't try to calculate sums if something turns hw accel back on
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This is due to hardware offloads - packets do not have correct checksums with the expectation the hardware will fill it in.
(Note this problem is already solved within meshnet; following that example.)
Disable Hardware Offloads via ethtool: When NewSocketHandler opens an interface, it uses github.com/safchain/ethtool (matching meshnet's approach) to disable active TX checksum and segmentation offloads as well as RX GRO/LRO/checksum offloads so the Linux kernel veth driver does not coalesce incoming segments or leave local frames with partial checksums.
Finalize Partial Checksums via PACKET_AUXDATA: Enabled PACKET_AUXDATA on the raw AF_PACKET socket and updated ReadPacket to inspect TP_STATUS_CSUMNOTREADY (skb->ip_summed == CHECKSUM_PARTIAL). If a frame arrives from a peer veth in another container namespace where TX checksum offload was still enabled, finalizePartialChecksum computes the missing L4 (TCP/UDP/ICMPv6) checksum in software before forwarding.
Unit Tests: Added unit tests covering offload disabling, PACKET_AUXDATA parsing, and IPv4/IPv6/VLAN partial checksum finalization.