Record what wrapping saved, and the rounding question left open

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>

docs/HANDOFF.md

@@ -8,7 +8,7 @@ log.
 
 Canonical Iris `main` is **`ca2b4b2`** (#17, the headless rig). **#18
 `split/18-position-chain`** is open in `/home/bob/repos/iris-pr18`; its local
-head is **`394d514`**, seventy commits. Built-in alignment is complete there;
+head is **`4febabf`**, seventy-one commits. Built-in alignment is complete there;
 see "Built-in alignment" below for the retained-layout details. No PR reviews
 were present when checked on 2026-09-15.
 
@@ -91,6 +91,11 @@ pin those rules. Seeds 10 and 86 are now in the ordinary generated set.
 
 ## Verification at `394d514`
 
+`4febabf` on top of it -- `Fixed` wrapping rather than saturating -- passed
+the workspace tests with all features (105), the release 100-seed oracle in
+9.5 s, and rendered `random`, `tabs` and `text` byte-identical at 1920x1200
+against `394d514`. The measurements are under "Performance".
+
 - `cargo fmt --all --check`
 - `cargo build --workspace --all-features`
 - `cargo clippy --workspace --all-targets --all-features -- -D warnings`
@@ -160,9 +165,14 @@ Measured on the fixed-shape fixture, seed 1, depth 8, 500 frames of `many`:
 | --- | ---: | ---: | ---: | ---: |
 | `5ed9e87`, before fixed point | 100 | 4,272 | 1,761M | ~699M |
 | `4cbb242` | 97 | 3,951 | 2,093M | ~819M |
+| `4febabf`, wrapping | 97 | 3,951 | 1,918M | ~771M |
 
-So the grid costs something like a fifth to a quarter, on a head that draws
+Normalised by primitive writes the grid is now about 1.18x the float head
-slightly less. Normalised by primitive writes it is about 1.27x.
+on instructions and 1.19x on cycles; per drawn widget it is 1.12x. What is
+left is the shape of an integer multiply with a rounding against a float
+fused multiply-add, spread across `UiSpan::within` (4.7% of cycles),
+`remap_subtree` and `Holds::through`; nothing in the profile above those
+is the grid. Text line breaking is 10% and `InstanceList::push` 9%.
 
 **Cycles are measurable on this machine and instructions alone are not
 enough.** Three runs of one binary varied 0.23%. It is *wall time* that
@@ -200,12 +210,22 @@ Tried and rejected, with numbers, so they are not tried again:
 - Short-circuiting `apply_scalar` where the fraction is nought or one: +17%.
   (From the previous session.)
 
-**The one lever left that is worth a decision:** `Fixed`'s `saturating_add`
+- `#[inline]` on `UiSpan::within`, whose prologue and epilogue were a sixth
-and `saturating_sub`, measured at **8.1% of instructions and 2.7% of
+  of its own cycles: -0.2% instructions, **+1.5% cycles**.
-cycles**. Wrapping instead is one instruction where saturating is five. That
+
-trades away "a clamped coordinate keeps the ordering a wrapped one inverts",
+**`4febabf` makes `Fixed` wrap rather than saturate** (Bryan, 2026-09-16: a
-which is a posture rather than a micro-optimisation, so it needs Bryan.
+coordinate past the range is not going to draw reasonably anyway, so wrap
-Everything else is spread thin.
+and break clearly): -8.6% instructions, -6.6% cycles, more on cycles than
+the 2.7% the instruction share predicted. `from_f32` still clamps and
+`Holds` keeps its saturating `narrow`, since a range of box lengths past
+`i32` really is unbounded.
+
+**One posture question is still open, with its number.** `shift_round`
+rounds halves away from zero so a value and its negation round to the same
+distance, which is a sign branch on every multiply. Rounding halves up
+instead measured **-1.6% instructions, -2% cycles** (1,888M, ~757M), and
+would let a flipped span land a step from its mirror image at exact halves.
+Not applied; Bryan's call.
 
 Earlier, against #18's own history: the retained rewrite took `many` from
 25.17M instructions a frame at `691e3eb` to 6.14M at `29c7881`, and `resize`
@@ -417,8 +437,12 @@ pixels and `Holds`. Decided with Bryan on 2026-09-15.
 - A weight is not a fraction: a list divides its room by the total of its
   weights, so `Weight` trades precision for the range to hold a whole list,
   and `Rel::ratio` turns two weights into a share on the finer grid.
-- **Arithmetic saturates rather than wrapping**, because a clamped coordinate
+- **Arithmetic wraps, as of `4febabf`.** It saturated first, so that a
-  keeps the ordering a wrapped one inverts.
+  clamped coordinate kept the ordering a wrapped one inverts, and that cost
+  a twelfth of layout's instructions to keep order two million pixels out
+  where nothing draws. `MIN` and `MAX` still stand in for an unbounded end,
+  which is safe only because every use compares against them and none adds
+  to them; `from_f32` is the one operation that clamps.
 - `Px` was `1/64` first. The residue of a length reached two ways is one
   rounding, so it scales with the step: at `1/64` that was 0.016 px, enough
   to move a box, and at `1/1024` it is a thousandth of a pixel. Range is