accept-aarch64 failed at BIRTH. The new CPU snapshot showed the kernel looping, not hung. The other stages passed, including riscv64, the stick, test-32, the proofs and the images. I ran them before reading aarch64's result, which was the wrong order. The repro boot (logs/20261119-012050) gave the code at each PC. Matched in the loader EFI's map, they name bitmap_get inside stadium_resident_heat_total, then background_decay and isqrt64. That is the heartbeat's load, which walked all 335,544 Stadium cells every tick. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_017RzZgrf7V7Zk2iczBXDRbm
12 lines
514 B
Markdown
12 lines
514 B
Markdown
# EMERGENCY diagnostic build -- evidence, not a result
|
|
|
|
After: `accept-aarch64` failed (that failure is the run's result). Commit: 36b258054cf58e8eb82bffb464725486a9acb15f.
|
|
Built with EMERGENCY_CONSOLE_ENABLED=1; nothing was booted (build only, ruled 2026-11-17).
|
|
|
|
To launch it by hand, with diagnostic logging, one QEMU at a time, in the foreground:
|
|
|
|
timeout 570 make -f kernel/Makefile ARCH=aarch64 EMERGENCY_CONSOLE_ENABLED=1 KERNEL_ARGS=--log-level=debug qemu
|
|
|
|
| Architecture | Build | Log |
|
|
|---|---|---|
|