[roll] Roll fuchsia [netstack3] Cache direction in per-packet filter metadata Currently, both conntrack and NAT logic independently attempt to determine the direction of a packet with respect to its conntrack connection. Because this check is stateless (i.e. is performed again at every NAT hook), if a packet is rewritten due to NAT, it is likely to no longer match either the original or reply tuple of its connection. This results in the netstack being unable to correctly perform NAT on it. Instead of recalculating it each time, have conntrack return the direction along with the connection itself when a packet's connection is looked up, and then stash it in the per-packet metadata that persists throughout the IP layer. This allows NAT to correctly be performed in all hooks that a packet traverses, and also allows slightly simplifying some conntrack code. Original-Fixed: 371017876 Test: netstack3-filter-test Original-Reviewed-on: https://fuchsia-review.googlesource.com/c/fuchsia/+/1168458 Original-Revision: f1fbd4f9027eb267d785d9c1cc13d4afb91410ae GitOrigin-RevId: 0ecf97942b63c7dda0a3c553687d1072117775eb Change-Id: I7a75a07a2d4933d181bd8a247424b3b7646ae681
This repository contains Fuchsia's Global Integration manifest files.
All changes should be made to the internal version of this repository. Our infrastructure automatically updates this version when the internal one changes.
Currently all changes must be made by a Google employee. Non-Google employees wishing to make a change can ask for assistance in one of the communication channels documented at get involved.
First install Jiri.
Next run:
$ jiri init $ jiri import minimal https://fuchsia.googlesource.com/integration $ jiri update
Third party projects should have their own subdirectory in ./third_party.