8d7721c775
B7A: QMI DMS time vs VNIIFTRI NTP (DMS = UTC - 0.28 s, sigma 1 ms; NITZ = DMS truncated to 1 s). B7B: a 1970 -> 2026 date -s step is harmless for the modem stack. B7C: aurora-modem probes DMS/NITZ along the start flow and steps CLOCK_REALTIME once at the first valid sample (REG registered/attached), never RTC; background SNTP check (query only) after START OK. 3/3 class-A cold boots PASS (residual +0.04/+0.36/+0.84 s), cache 14fe453a written (B7C_WRITE_VERIFIED) and verified by a normal power-on (V1, +0.26 s). Also publishes the prerequisite R2 (boot-hang / lk eMMC investigation, T4 telnet baseline that B7C builds on) and R3 (autonomous cold boot, NCM loss) material, and extends tools/publish-sanitize.py to r2/, r3/, b7/. Binary images, initramfs, busybox and raw logs stay out (see b7/*/SHA256SUMS).
687 B
687 B
B7C c4b — functional run, class B (operator: "грузи щас"; booted from the fastboot left by c4 start 3; RTC 318 s). Not counted.
TIME SET at uptime 21.26 s, phase REG registered/attached, DMS−NITZ +0.24 s; START OK 73.0 s; SNTP check +0.580 s (best, ntp3), verify ntp2 +0.586…+0.592 s; clock kept (< 1 s), UNSYNC kept, RTC untouched, eMMC writes 0, dmesg-errs 0, LTE ping 3/3. Residual over the functional runs: c1 +0.19 s, c2b +0.45 s, c4b +0.58 s (B7A on a long-running modem: +0.28 s, σ 1 ms). Hypothesis: right after registration DMS is set from whole-second NITZ, so its error is a per-boot value in [0, 1) s. Within the B7C limit (0.3–1 s), but not a fixed bias.