commit | 1cb2d6c7acf6e37cb1462f271de6b7690f259979 | [log] [tgz] |
---|---|---|
author | Filip Filmar <fmil@fuchsia.infra.roller.google.com> | Fri Feb 14 10:51:24 2025 -0800 |
committer | Copybara-Service <copybara-worker@google.com> | Fri Feb 14 10:55:22 2025 -0800 |
tree | 5e6c6d9e307816577673c3c14150945703a79aa4 | |
parent | 5cfcf9eba68954c32dcb59ff634cb5d2edcdaa2a [diff] |
[roll] Roll fuchsia [starnix][hrtimer] Handles `setup_done` I wanted to remove this from the list of TODOs which may affect the correctness of the code. For quite a while we did not wait for the `setup_done` event, which had the potential to mess up the collaboration with the wake alarm manager. List of changes: - Handling `setup_done` signal. The "start timer" now waits until the wake alarm manager notifies that the alarm has been scheduled, before proceeding - Subtle change of the ACK (K=1) semantics which is now required to avoid deadlocks. We now signal K=1 as soon as we get a FIDL message, instead on relying on iteration. It adds more event signaling into the code path which in theory has performance implications. Not sure if this is something we should worry about now. - Plentiful tracing annotations. Those are a lifesaver! - Re-did the test fake for the wake alarm manager to use its own thread(s) to run. This now seems necessary because otherwise the lock uses in the test code interferes with the ability to run the fake FIDL server. Since this scenario is impossible in production, given that the wake alarm manager is its own process, this seemed like a correct thing to do. Multiply: starnix-tests Original-Bug: 373731551, 373928684 Original-Reviewed-on: https://fuchsia-review.googlesource.com/c/fuchsia/+/1193236 Original-Revision: 004181940d20ba7b14010baea9311feef1b429de GitOrigin-RevId: 24aaa3c837b07dbeb9c4f7310b95d6f8e820cd68 Change-Id: I4d4e0e1441c47235d311d191d3669c359f30fcbe
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
.