[roll] Roll fuchsia [virtualization] Wire in completion queues These queues batch completions (without actually delaying the completions -- batches are accumulated until the completion task runs). We're (probably temporarily) measuring queue statistics so that we can tweak the queue depths. This has to be moved out of the destructor so that it can be run before the dispatch loop shuts down so that logging actually happens. Multiply: virtualization_net_tests Original-Bug: 95485 Tested: Used iperf3 from VM -> gLinux workstation, saw an average completion queue utilization of 5% ~ 12 completions were batched (and thus latency was improved for the 11 which would have been scheduled separately) Net throughput looks similar which we'd expect, but variance due to bridging makes it hard to say. Original-Reviewed-on: https://fuchsia-review.googlesource.com/c/fuchsia/+/751782 Original-Revision: 303333108d8cbaf1f2209f01c2fa5995076dbff7 GitOrigin-RevId: 1cb724dc715e48d054ef8988c78237dadaed5092 Change-Id: I390515d277a06f9402a095f5dd84ad771bd9aeb9
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 via the IRC channel #fuchsia on Freenode.
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.