# Session 2026-10-22 — corpus triage, libc, SYSTEM, proc types, sized storage, SysClock, small fixes

Branch **`master`**.  Continues `summary_session-2026-10-19.md`; the
only plan source was its "Remaining" list (corpus `LANG` triage).

## Numbers

| gate | session start | end |
| --- | --- | --- |
| suite (`compiler/run_tests.sh`) | 249/249 | **259/259** |
| fixpoint (`bootstrap/fixpoint.sh`) | OK (3,882,337 B) | **OK (4,028,447 B)** |
| corpus (`tools/v3-corpus/corpus.sh`) | 467/612 | **511/612, 0 crashes** |

Reproduce:

```sh
cd compiler && ./build.sh && ./run_tests.sh          # 259/259
cd .. && TMP="$PWD/.scratch/fixpoint" ./bootstrap/fixpoint.sh
tools/v3-corpus/corpus.sh                            # 511 OK, 0 crash
```

## Triage method

The 118 `LANG` entries were compiled one at a time (a single-file
version of `corpus.sh`), and each first diagnostic was read against its
source line.  They fall into these buckets:

| bucket | count | verdict |
| --- | --- | --- |
| imports `libc` but uses a libc function as a value (`printf`/`malloc`) | ~48 | libc prototypes were missing; **fixed below** (the rest fail later for other reasons) |
| declaration order (type/const/var used before its declaration) | ~25 | correctly rejected (`docs/gaps/`); not a target |
| gm2 extensions (`ASM`, `''` NUL char, `666B`, absolute `VAR [addr]`) | ~7 | not a target unless chosen |
| ISO features actually missing | ~20 | `PACKEDSET`, exceptions/`UNIQUE`, `__FUNCTION__`, `PROC` as a result type; sized SYSTEM types **fixed below** |
| **conforming** clusters: arrays, subranges, ordinal `FOR`, empty string, libc prototypes, SYSTEM sized types, procedure types, SysClock, small fixes | 44 | **fixed this session** |

## What landed

### PIM/ISO multi-index arrays — `compiler/src/M2.atg`
`ARRAY [0..1], [0..2] OF T` is the standard spelling (17 corpus files
use it; none use V3's single-bracket comma list).  `ArrayType` now
parses one bracket group per dimension (`{ "," "[" ... "]" }`) and
appends every group's bounds before the element type; nesting stays
inside-out, so the layout matches `ARRAY [0..1] OF ARRAY [0..2] OF T`.
The old `ARRAY [0..1, 0..2] OF` form still works.  Corpus **+6**:
`arraybool`, `assignment2`, `foo`, `recordarray2`, `tinyarray4/5`.
Regression `t_multidim`.

### Subranges keep their ordinal base — `SymTab`, grammar
A subrange was always `FSub` with no base descriptor, so `ClassOf`
returned `ClInt` for `['A'..'Z']` and `[red..blue]`: assigning a
`CHAR`/enum literal (or comparing with one) failed.  `SymTab.NewSubB`
stores the base when both bounds have the same type; `ClassOf`,
`SubBounds`, `SetSpan`/`BaseSpanOk` and the set-span check now follow
it.  `Assignable`/`EqCheck`/`OrdCheck` accept a base value against a
subrange of the same `CHAR`/`BOOLEAN`/enum base (two different enums
stay incompatible).  Corpus **+3**: `subrange16`, `subrange17`,
`enum3`.  Regression `t_subrange`.  Assignment range checks remain
(future) — V3 is already lenient for the integer family.

### `FOR` over ordinals + typed `MIN`/`MAX` of sets — grammar
`FOR` accepted only the integer family.  It now accepts any ordinal
control variable (bounds must be assignment-compatible with it), which
covers enums and `CHAR`.  `MIN`/`MAX` of a `SET` now yields the set's
base type instead of `INTEGER`, so `FOR i := MIN(S) TO MAX(S)` binds
(the bound was previously `INTEGER` against an enum `i`).  Corpus
**+3**: `forloop2`, `smallset4`, `minmax`.

### ISO empty string literal — `SymTab` + grammar + `Lower`
ISO makes the empty string literal (`""` / `''`) length 0 and
compatible with `CHAR`, where its value is the terminator.  V3 typed it
as an ordinary string, so `ch := ""`, `ch = ""`, `CASE ch OF ""`,
`FOR ch := "" TO …`, a `CHAR` formal actual, and `RETURN ""` either
failed (`incompatible assignment`/`comparison`, `invalid Fact`) or
stored a descriptor address.

`SymTab.NewEmptyStr` marks the literal's type (`tlo = -2`); it is still
`ClStr`, so `ARRAY OF CHAR` contexts keep working (`s := ""`, `s = ""`,
`t_strempty`).  `Assignable`/`EqCheck` accept it against `CHAR`, and
`Lower` substitutes the value `0` when the context is a character
(assignment, relation, `CASE` label, `FOR` bound, value actual,
`RETURN`).  Corpus **+6**: `arraychar`, `arraychar2`, `arraychar3`,
`nulcharcase`, `nulcharvar`, `forloopnulchar`.  Regression
`t_emptychar`.

### `libc` prototypes — `SymTab` + `Lower` + grammar
`FROM libc IMPORT P` used to declare every name as a permissive
no-result varargs external, so `printf`/`malloc`/`write` in an
expression were `invalid call`.  The known C functions now get real
signatures from a built-in table (`DeclareCProc`/`CProto`):
`printf`/`snprintf` are varargs functions with an `ARRAY OF CHAR`
format, the rest have typed formals/results (`exit` none, `write`/
`read`/`open`/`close`/`lseek`, `malloc`, `memcpy`/`memset`/`strncpy`,
`system`/`getenv`, `atof`/`strtod`).  `Lower.NoteCProc` copies the
signature into its proc table.

Discarding the result of a call in statement position is now allowed
**only for foreign C varargs** (`printf`/`snprintf`) — the documented
compatibility carve-out.  An ordinary Modula-2 function used as a
statement is still `233` (`t_bad_funstmt`).  Corpus **+6**:
`procedure2`, `tinyparam`, `tinyparam2`, `sizes`, `variant9`,
`varient` (`except5` now stops on ISO exceptions).  Regression
`t_libc2`.

### ISO procedure-type syntax — `M2.atg`
`PROCEDURE` types accepted `;`-separated sections and the GNU type-only
`(T1, T2)` shorthand, but not the ISO/PIM forms used by the corpus:

- comma-separated unnamed parameter types with a keyword/`VAR` start:
  `PROCEDURE(ARRAY OF CHAR, ARRAY OF CHAR)` and
  `PROCEDURE (CHAR, VAR scanClass, VAR scanState)`;
- `PROC` as a result type (`PROCEDURE ... : PROC`), since `PROC` is a
  keyword token and the result position only took `TypeIdent`.

The parameter list now parses a comma as either another name or a new
unnamed type entry; if a type entry starts, the collected names are
resolved as types (`CHAR`, a declared type, or the GNU shorthand).  The
name-group behaviour (`a, b : T`) is unchanged.  `PROC` in a result
position maps to the predeclared procedure-value type.  Corpus **+4**:
`typeonly.def`, `testrecursive.def`, `procconv`, `procconv2`.
Regression `t_proctype2`.

### True 8-bit `BYTE`/`SBYTE` — `SymTab` + `QbeGen` + `Lower`
`SYSTEM.BYTE` had been an alias for the 32-bit `CARDINAL` (a phase-1
shortcut: only `w` integer values existed), so `SIZE(BYTE)` was 4 and
`ARRAY OF BYTE` was not byte-granular.  `BYTE`/`SBYTE` (and `WORD8`/
`CARDINAL8`/`INTEGER8`) are now distinct 8-bit storage types:
`TypeSize`/`ElemBytes` 1, `loadub`/`loadsb`/`storeb` for scalars,
parameters, record fields and array elements, `{ b 0 }` for globals
and static data, and arithmetic promotes to `INTEGER`/`CARDINAL`.
The `gap09` scalar→`ARRAY OF BYTE` wrapper now copies one byte for a
`BYTE` actual; actual arrays are passed as real byte buffers.

`WORD16`/`WORD32` (and the 16-bit named types) still map to the 32-bit
word model, and `COMPLEX*` remains a `REAL` placeholder.  Regression
`t_byte` (SIZE, wrapping, signed loads, record/array fields,
`ARRAY OF BYTE`, `SET OF BYTE`, `LONGINT`/`FLOAT`/`VAL`).  Corpus
unchanged at 503 (no file depended on the old 32-bit `BYTE`).

### True 16-bit `WORD16`/`CARDINAL16`/`INTEGER16` — same technique
The 16-bit names were aliases of the 32-bit family.  They now use
dedicated forms/classes (signed/unsigned) with `SIZE` 2,
`loaduh`/`loadsh`/`storeh` in scalars, parameters, record fields and
array elements, `h 0` data for globals/static aggregates, and
arithmetic promoting to `INTEGER`/`CARDINAL` (stores truncate to 16
bits).  `SET OF WORD16`/`SET OF INTEGER16` work (65536-element span,
the documented set limit).  `WORD`/`WORD32` remain the 32-bit word.
Regression `t_word16`.

### ISO `SysClock` — `stdlib/sysclock.def` / `.mod`
The stdlib stub only had PIM-style `Time`/`Clock`, so `testclock` and
`testclock2` failed on `CanGetClock`/`DateTime`/`GetClock`.  `SysClock`
now implements the ISO 10514-1 interface: `maxSecondParts`, `Month`/
`Day`/`Hour`/`Min`/`Sec`/`Fraction`/`UTCDiff`, `DateTime`,
`CanGetClock`, `CanSetClock`, `IsValidDateTime`, `GetClock`,
`SetClock`.  `GetClock` converts the shim's epoch seconds to a UTC
civil date (Hinnant's algorithm; `zone = 0`, no DST); `SetClock` is a
no-op with `CanSetClock = FALSE`.  `Time`/`Clock` remain as extras.
Corpus **+2**: `testclock`, `testclock2`.  Regression `TSysClock`.

### Small conformance items — grammar + `Lower`
Five corpus gaps plus one incidental:

- one-character string literals concatenate (`"W" + "o"`): both are
  `CHAR`-typed immediates, so the fold builds the quoted 2-character
  literal and the AST stays literal; longer `CONST`s then fold through
  the existing string rules;
- ordinal relations of constants fold in `Expr` (needed by
  `ORD(a > b)` in a `CONST`);
- `HIGH`/`LEN` of a string `CONST` use `QbeGen.StrLitLen` (the pooled
  literal text), in both the parser's constant folding and `Lower`;
- `INC`/`DEC` accept any ordinal (`CASE`/enum control), not just the
  integer family;
- a leading `|` before the first `CASE` alternative is tolerated;
- `__FUNCTION__` is a predefined literal folded to the current
  procedure name (module name at unit level; `unitName` is now set for
  program modules too).

Corpus **+6** (`constbool4`, `hexstring`, `function`, `enum2`,
`testcase`, plus one incidental).  Regression `t_smallgaps`.

## Commits

Three tagged states, each green on all gates:

| commit | tag | scope |
| --- | --- | --- |
| `aa4f49b` | `v3-corpus-triage` | multi-index arrays, typed subranges, ordinal `FOR` (+12) |
| `f87bcc9` | `v3-empty-string` | ISO empty string as `CHAR` (+6) |
| `c02fd41` | `v3-libc-prototypes` | typed libc prototypes, `printf` varargs (+6) |
| `713650f` | `v3-system-types` | ISO SYSTEM sized types + qualified `SYSTEM` (+8) |
| `f236600` | `v3-proc-types` | ISO procedure-type syntax (+4) |
| `7621f16` | `v3-byte` | true 8-bit `BYTE`/`SBYTE` storage |
| `a5d0f16` | `v3-word16` | true 16-bit `WORD16`/`CARDINAL16`/`INTEGER16` |
| `4d05afe` | `v3-sysclock` | ISO `SysClock` API |
| `87e8605` | `v3-smallfix` | small conformance items (+6) |

## Lessons

- A corpus "bucket" can be misleading until each file's **first**
  diagnostic is read: 48 files import `libc`, but most stop earlier on
  declaration order, exceptions or sized types.  Only 6 were blocked by
  the missing prototypes.
- The `printf` decision was a policy choice, not a technical one: the
  permissive external already made statement calls work.  Option 3
  (function + allow discarding the result only for foreign C varargs)
  keeps `t_bad_funstmt` and unlocks all 59 `printf` files.
- `docs/gaps/README.md` is still the place to record *why* a case is
  rejected; the new entries state the fix and the remaining carve-out.

### ISO SYSTEM sized types + qualified `SYSTEM` — `SymTab` + grammar
The hard-coded `SYSTEM` module was not registered, so
`SYSTEM.CARDINAL8` (qualified type access) was undeclared even though
`FROM SYSTEM IMPORT CARDINAL8` worked.  `InitSystemModule` now registers
a pseudo-module with its own scope and copies every predef type and
constant into it; `TypeIdent`'s `QualType` and `DesignTail`'s
`MaterializeAlias` then resolve `SYSTEM.X` normally.  `FROM SYSTEM
IMPORT` stays permissive (its names are global predefs or keywords).

Added the ISO/gm2 names: `WORD8/16/32/64`, `BITSET8/16/32`,
`REAL32/64/128`, `COMPLEX32/64/128`, `CSIZE_T`/`CSSIZE_T`/`COFF_T`,
and `BITSPERLOC`/`LOCSPERWORD`/`LOCSPERBYTE`.  At this step V3 still
had a 32-bit word model, so the narrow integer/word names shared
`INTEGER`/`CARDINAL`; later in the session `BYTE`/`SYSTEM` 8-bit names
and `WORD16`/`CARDINAL16`/`INTEGER16` became true 8/16-bit storage (see
below).  `COMPLEX*` remains a `REAL` placeholder.  Corpus **+8**:
`compsize`, `int8field`, `returnrecord`, `printsize`, `testgeneric`,
`testgeneric2/3/4`.  Regression `t_system`.

## Remaining (for a next session)

1. **A real `libc.def`** — the built-in prototype table covers the
   common functions; a definition module (or `DEFINITION MODULE FOR
   "C"`) would let users extend it.  Not needed for the corpus now.
2. **`PACKEDSET`**, ISO exceptions/`UNIQUE` — each one or two files;
   larger or gm2-flavoured.
3. **Runtime subrange range checks** — the remaining conformance
   difference from this session's subrange work.
4. `STATUS.md` "what remains" is the live list.

## Publication

- Branch `master`; feature tags `v3-corpus-triage`, `v3-empty-string`,
  `v3-libc-prototypes`, `v3-system-types`, `v3-proc-types`, `v3-byte`,
  `v3-word16`, `v3-sysclock`, `v3-smallfix`.
- `v3-session-2026-10-22` is a **moving** session tag: it was advanced
  after publication, so the remote rejects updating it without
  `--force`.  The feature tags are immutable checkpoints.

```sh
git push origin master v3-smallfix
# only if you want the moving tag advanced on the remote too:
git push --force origin refs/tags/v3-session-2026-10-22
```

## Stop point

Session stopped here by request.  Final gates: suite **259/259**,
fixpoint **OK (4,028,447 bytes)**, corpus **511/612, 0 crashes**.
Feature tag `v3-smallfix`; the session summary is tagged
`v3-session-2026-10-22-end`.  The next-session plan is
`docs/handoff-2026-10-22-next.md`.