BlackMark
74d8b92885
The second pass concluded the 168 B tricks->asm gap was per-call ABI
cost. Most of it was structure. Rebuilt around the oracle's own shape —
argless noinline primitives over a whole-loader call-saved register
protocol (g_addr in Y, count r16, window r7, direction latch r6), a
top-down erase_below whose loop tests against zero and hands callers
g_addr = 0 for free, bounded rx everywhere (a silent host unwinds to
the app from any state, as the oracle does), and a named tsb_app entry
that --pmem-wrap-around=32k relaxes to the wrapped rjmp:
tsb_asm 510 B in the 512 B section (oracle: 500), C++ except rx
and the page-store loop — the two routines whose remaining
cost is the calling convention itself (~30 asm lines, was
~280)
tsb_tricks 526 B, no assembly at all (was 666)
tsb_pure 836 B, still one readable function per command (was 842)
Every g_* update placement works around a GCC 16.1 wrong-code bug
(stores into global register variables deleted when only callees read
them — repro and rules in libavr dev/lessons.md). Also fixes two
latent hardware bugs all earlier tiers carried, masked by simavr's
zeroed register file: the crt-less entries never established
__zero_reg__ = 0, and the direction latch was read before written —
power-on registers are undefined.
All tiers full oracle feature parity, protocol tests green in both
libavr modes, .text byte-identical across modes.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Oracle — the hand-written TinySafeBoot assembly
tsb-fixedbaud.asm is the reference implementation this port is measured
against: the native-UART, fixed-baud TinySafeBoot bootloader, hand-written
in AVR assembly. It is the size-and-feature bar for the port's tsb_asm tier.
- Source: https://github.com/seedrobotics/tinysafeboot
(
firmware_ASM/latest_stable_release/20200727-fixedbaud/main.asm), the Seed Robotics fixed-baud fork of Julien Thomas' TinySafeBoot. - License: GPLv3 (see the header in the file). It is vendored here only as a reference oracle — it is not compiled, linked, or distributed as part of the MIT-licensed port. Mere aggregation.
Why this variant
The user chose the fixed-baud, hardware-UART variant deliberately: it is the one whose feature set the port must match. It fits the complete TSB feature set into the 512-byte ATmega boot section:
| Feature | Oracle routine |
|---|---|
| Watchdog-reset bail straight to the app | RESET (WDRF check) |
| One-wire half-duplex (RX/TX shorted): RXEN/TXEN toggled per direction, TX turnaround guard | SetRX / SetTX / TransmitByte |
| Activation timeout read from the config page, with a lockout-proof minimum | WRX1To (uses utimeoutH) |
3×@ activation knock |
ActCharRcvd |
| Password gate; wrong byte hangs (still draining the UART) | CheckPassword |
Emergency erase on password \0 + double-confirm — wipes flash, EEPROM and the config page |
EmergencyErase |
| Device-info block (16 bytes) | SendDeviceInfo / DEVICEINFO |
App-flash read/write (f/F), EEPROM read/write (e/E), config read/write (c/C) |
CheckCommands |
Assembled size (the bar)
Assembled for the ATmega328P with avra:
avra -I /usr/share/avra tsb-fixedbaud.asm # after uncommenting .include "m328Pdef.inc"
# Code : 250 words (500 bytes) — the whole loader, all features, in the 512 B section
500 bytes with every feature — the proof that ≤512 B and full feature parity
are simultaneously reachable. The port's tsb_asm tier meets the same bar at
510 B in the same 512 B section, written in C++ on libavr except the two
routines whose remaining cost is the calling convention itself (the bounded rx
and the page-store loop); tsb_tricks needs no assembly at all at 526 B, and
tsb_pure stays fully idiomatic at 836 B, both in the 1 KB section.
The oracle targets 20 MHz / 33333 baud; the port targets 16 MHz / 115200 baud (what the simavr protocol test drives). Baud and geometry differ, code size and feature set do not.