| Filename | Latest commit message | Latest commit date |
|---|---|---|
208ffd374f
Some checks failed
energy meters on real hardware / sink decode table · per-task energy and the GPU rail on macOS (push) Successful in 5m13s
repository checks / lockfiles, advisories, tests in CI, workflow hygiene, spec and section references, public text (push) Has been cancelled
energy meters on real hardware / RAPL and NVML on jetson-hub (push) Failing after 2m37s
jouleclaw public surface must stay documented / Public-surface audit (jouleclaw-rs) (push) Failing after 15m17s
energy meters on real hardware / sink decode table · RAPL on jetson-hub (NVML surveyed) (push) Failing after 20s
The docs of each crate's IoReportStatus still stated the liveness rule as it stood before the cap: five advances, four in five of the intervals observed, 300 ms of wall time. Live, and Frozen where it states the rule, now say it as the spec texts do: what an interval is, that only one no longer than three of the probe's ticks (75 ms at its 25 ms tick) counts, that a longer one counts as nothing, and that the 300 ms is the summed length of the intervals that count. Frozen also says that a probe the host held, so that too few of its intervals counted, reads Frozen. In eoc-meter, jcp-energy, jouleclaw-energy, proof-cascade and sandbox-meter. Doc comments only. Changelog: none Co-Authored-By: Claude Opus 5.5 <[email protected]> |
||
| .. | ||
| proof-cascade | ||
| proof-cli | ||
| proof-conformance | ||
| proof-receipt | ||
| proof-schema | ||
| proof-solvers | ||