docs: what the 64-bit millisecond clock costs, measured

Two findings from the flash sweep, both measured and neither taken, because
each is a decision rather than a defect.

Narrowing uptime::millis() and its three timestamps to 32 bits is 608 B, 7.6 %
of the image, and 16 B of RAM. The reason it is not a free win is not the
display: a 32-bit millisecond counter wraps every 49.7 days on a board that
runs continuously, and the interval tests would have to become subtractions,
since `now - last >= interval` survives a wrap where `now >= last + interval`
does not - which is how terminal.hpp's monitor tick is written today. The
64-bit counter is what puts the wrap out of reach, so this is a question about
a wrapping uptime display, and the answer decides 608 B.

The 1 kHz tick's own prologue is the smaller one. The handler increments 64
bits, so it calls __adddi3_s8, and a call in a signal handler decides the
prologue - twelve push/pop pairs for what the helper might clobber. Two 32-bit
halves remove the call and keep the range, taking the handler from ~47
instructions to ~26 with the carry running once every 49.7 days. Not taken
either: it costs 66 B of flash to buy about 0.4 % of the CPU, and nothing here
is timing-critical.

This repo had nowhere to record work, so it has a tracker now.

Docs only; the image is unchanged at 8004 B and the suite is green.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

dev/tasks.md

@@ -0,0 +1,39 @@
+# Tasks
+
+This file is current work. The repo is the sole task tracker (libavr guidance
+rule 20).
+
+## Open
+
+- [ ] decide whether the millisecond clock stays 64 bits. **Measured: 608 B of
+      flash, 7.6 % of the image**, plus 16 B of RAM - `uptime::millis()` and the
+      three timestamps that hold its value narrowed from `uint64_t` to
+      `uint32_t`, 8004 B down to 7396.
+
+      It is not free, and the price is not only the display. A 32-bit
+      millisecond counter wraps every **49.7 days**, and this board runs
+      continuously, so two things follow. The `uptime` command would restart
+      from zero at each wrap. And the interval tests would have to be rewritten
+      as subtractions - `now - last >= interval` is correct across a wrap where
+      `now >= last + interval` is not, and `terminal.hpp`'s monitor tick is
+      written the second way today. That shape is *why* the counter is 64 bits:
+      it puts the wrap out of reach so a naive comparison cannot be wrong.
+
+      So the question is whether a wrapping uptime display is acceptable, and
+      the answer decides 608 B. If it is, the three call sites in
+      `statistics.hpp` and `terminal.hpp` move to subtraction in the same
+      change, and that is the whole of the work.
+
+- [ ] the 1 kHz tick's own cost, which is separate and smaller. The compare
+      handler increments a 64-bit counter, so it calls libgcc's `__adddi3_s8`,
+      and a call inside a signal handler decides the prologue - twelve push/pop
+      pairs to cover what the helper might clobber. Splitting the counter into
+      two 32-bit halves removes the call and halves the prologue: the handler
+      goes from ~47 instructions to ~26 and keeps the full 64-bit range, the
+      carry running once every 49.7 days.
+
+      **Measured and not taken, because it costs 66 B of flash to save about
+      0.4 % of the CPU** - `millis()` then reassembles the halves, and the
+      readers pay for it. It is the right change only if the tick's cycles ever
+      matter; today nothing here is timing-critical. Falls away entirely if the
+      counter narrows above, which is the reason to decide that one first.