lab/sx
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: resolve qualified-import-member const as a compile-time constant (issue 0192)

Browse Source 
A namespaced import's const (`m :: #import "lib.sx"; … m.CAP`) only ever
resolved as a runtime value — the const folders in program_index.zig had no
namespace-member arm, so a qualified const was rejected as an array dimension /
Vector lane / generic value-param and could not seed another const, while the
flat-import form worked everywhere.

Add a `lookupQualifiedConst` (+ float / float-typed twins) ctx hook: resolve
the alias via `namespaceAliasVerdictFrom` to its target module, then fold the
member from that module's per-source const cache (`foldQualifiedConstInt` in
lower/comptime.zig), pinned to the target source so nested const RHSs fold
there. Wire it into evalConstIntExpr / evalConstFloatExpr / isFloatValuedExpr —
both the expression-position field_access arm (`[m.CAP]T`) and the
type-argument dotted-name arm (`Vector(m.LANES, …)`, generic value-params).

Implemented on the source-aware ctxs (Lowering / SourceConstCtx); the
namespace-blind ModuleConstCtx / StatelessInner return null, so a qualified-const
dim reached only via the stateless type-alias path stays a clean unresolved-dim
diagnostic, never a fabricated length. Resolves correctly for array dims,
arithmetic, integral-float dims, Vector lanes, generic value-params, inline-for
bounds, and struct fields.

Regression: examples/modules/0842-modules-qualified-import-const-comptime.sx.
This commit is contained in:
agra
2026-06-26 07:51:27 +03:00
parent 6b8bce1aba
commit 501399b1a9
12 changed files with 361 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
current/CHECKPOINT-FIBERS.md
View File

@@ -518,6 +518,32 @@ non-unification: virtual-time timers and real kqueue timeouts are NOT merged —
timer before ever blocking on kqueue (a program uses `sleep` OR fds); a true "fd-or-real-timeout" wants
a kqueue `EVFILT_TIMER`, future work.
> **✅ issue 0192 FIXED (2026-06-26) — epoll work UNBLOCKED.** A qualified-import-member const
> (`m.EV_SIZE`) now folds as a compile-time constant in every position the bare/flat form does
> (array dim, arithmetic, Vector lane, generic value-param, inline-for) — so the clean
> `[MAXEV * ep.EV_SIZE]u8` event buffer the bindings want will work. Fix: a `lookupQualifiedConst`
> ctx hook resolving the namespace alias → target module's per-source const, wired into the int/float
> const folders (`src/ir/program_index.zig` + `src/ir/lower/comptime.zig`). Regression:
> `examples/modules/0842-modules-qualified-import-const-comptime.sx`. The hint stands for the rebuild:
> **a struct-per-arch `EpollEvent` (arch-branched u32 fields, 12 B x86_64 / 16 B aarch64) beat raw
> byte access** — idiomatic field reads, no issue-0155 scalar-pointer indexing, no unaligned u64.
> Resume: rebuild `std/net/epoll.sx`, branch `std.event.Loop` on `inline if OS`, lock with a darwin run
> + ir-only linux example.
> **⛔ (HISTORICAL) BLOCKED on issue 0192 (filed 2026-06-26).** Started the epoll work: chose the `std.event.Loop`
> backend (pure sx + libc externs, zero compiler change — per "do this in sx as much as possible") as
> the first deliverable, since event.sx already names epoll as its linux backend and it's runnable
> (darwin via kqueue) + ir-only-verifiable (linux). De-risked four landmines by probe — arch-dependent
> layout const via module-scope `inline if ARCH` (folds + validates in linux IR), slice-based byte access
> (sidesteps issue 0155), no unaligned u64 (store the 32-bit fd in epoll `data`), and comptime-dead linux
> externs don't break the darwin corpus (just an unreferenced `declare`). Then hit a compiler bug while
> sizing the event buffer: a **qualified-import-member const is not a compile-time constant** —
> `[m.CAP]u8` / `A :: m.CAP` fail (a *flat*-imported const works). Root cause located:
> `evalConstIntExpr` (`src/ir/program_index.zig:325`) has no namespace-member-const arm. Per the STOP
> rule the half-built `std/net/epoll.sx` (which used a struct-based layout to route around the bug) was
> **removed**, not landed — the unblock session rebuilds it cleanly with the fix in hand. Repro +
> investigation prompt: `issues/0192-qualified-import-const-not-comptime.{md,sx}`.
Design note carried forward: an event-loop `Io` needs a current-`Scheduler` handle. `sched.*` methods
thread it via `self`/the `Task`; if B1.4c wants the capability-threaded `context.io` form it'll need
an ambient current-scheduler accessor in sched.sx (still deferred — the `sched.*`-method form