Skip to content

Document real peer frame timestamp semantics #8

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
glocker merged 1 commit into from
Nov 11, 2026
Merged

Document real peer frame timestamp semantics #8

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 16 additions & 0 deletions docs/v1-scope.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -691,6 +691,22 @@ warning: transport.type='socketcan' (interface 'vcan0') without --realtime; virt
Warnings go to `stderr`, so they compose with `--json`/`--quiet` without
corrupting machine-readable `stdout`.

#### Timestamps of frames from a real peer

Virtual time advances only at scheduled instants: the 1 s physical step and any
scenario events. A frame from a real peer that arrives *between* two of them is
stamped with the earlier one. This applies to every frame on every transport and to
everything derived from the stamp: the `t=` of its `--trace` line, of a `DROP` line
and of a `dropped_packets` entry in `--json`. So with `--realtime` such a frame can
read up to one physical step (1 s) earlier than it actually arrived; a bad frame
written 1.4 s into a wait is reported at `t=1000000`. Within one run the stamps are
consistent with each other (a frame and its drop share one), and order is preserved.

This is deliberate: stamping arrivals from the wall clock would make traces depend
on OS scheduling and weaken the guarantee that the same inputs give the same trace.
With the default virtual clock nothing arrives mid-wait, so none of this applies.
Treat `t=` on traffic from a real peer as "at or after this virtual instant".

Note that virtual-time durations now cost real time: a scenario that waits
`30s` takes 30 seconds under `--realtime`. Size scenario waits and assertion
timeouts accordingly.
Loading