Repository navigation

Issues

New issue

Umbrella: proof-mode e2e readiness — any valid program + any spec => dischargeable .v
#366 · 0xGeorgii opened on Aug 8, 2026
Configurable, statically-known linear-memory size for generated WASM (CLI flag + [wasm] Inference.toml)
#221 · 0xGeorgii opened on Jun 6, 2026

CHANGELOG: the #221 Breaking entry names ManifestSettings fields that do not exist
Inferara/inference#550 · opened on Oct 26, 2026
infs: the memory: build line shows the requested stack, not the one static data leaves
Inferara/inference#549 · opened on Oct 26, 2026
infs: re-check a defined memory's limits after [build.wasm-opt]
Inferara/inference#548 · opened on Oct 26, 2026
SpaceWasm builds with an imported linear memory
Inferara/inference#547 · opened on Oct 24, 2026
Link [wasm-dependencies] modules into a module that imports its memory
Inferara/inference#546 · opened on Oct 24, 2026
Module-scope constants (#515): follow-up tracker
Inferara/inference#537 · opened on Oct 20, 2026
Module constants: tests for guarantees nothing currently enforces
Inferara/inference#536 · opened on Oct 20, 2026
Module constants: documentation, diagnostic wording and CHANGELOG corrections
Inferara/inference#535 · opened on Oct 20, 2026
A module constant whose type is item-imported fails to type-check
Inferara/inference#534 · opened on Oct 20, 2026
Module constants in specifications: constant indices aren't folded, and tables over 64 elements can't be indexed symbolically
Inferara/inference#533 · opened on Oct 20, 2026
A036 and A058 suggest memory layouts that fail the other check or that the build refuses
Inferara/inference#532 · opened on Oct 20, 2026
Module constants: place static data per build mode and hand analysis and codegen one placed layout
Inferara/inference#531 · opened on Oct 20, 2026
Compute type layout once: codegen, A036 and the constant evaluator each have their own
Inferara/inference#530 · opened on Oct 20, 2026
Module constants: initializers skip most analysis rules, and constness depends on values
Inferara/inference#529 · opened on Oct 20, 2026
Module constants: the evaluator panics, aborts or emits an invalid module on user input
Inferara/inference#528 · opened on Oct 20, 2026
wasm-to-v: name-handling follow-ups from #405 (stdlib import capture, embedded module name, spec-name over-rejection, stale hints and docs)
Inferara/inference#518 · opened on Oct 6, 2026
continue; is reported as an undeclared variable, and compiles as a no-op when a binding named continue is in scope
Inferara/inference#505 · opened on Sep 29, 2026
Stellar: contractspecv0 has an empty doc for every method; /// doc comments never reach it
Inferara/inference#504 · opened on Sep 29, 2026
book: Stellar proving procedure says proofs/contract.wasm is the pre-rewrite module byte for byte, but a spec block makes it a superset
Inferara/inference#503 · opened on Sep 29, 2026
infs has no proof build for a stellar or spacewasm project without editing [build] target, and the refused proof build deletes the previous artifact
Inferara/inference#502 · opened on Sep 29, 2026
Overflow guards cost the ray-tracer benchmark 72% of its throughput
Inferara/inference#501 · opened on Sep 27, 2026
Allow bulk memory for SpaceWasm builds once nasa/spacewasm#54 ships
Inferara/inference#500 · opened on Sep 27, 2026
Check the SpaceWasm limits against upstream spacewasm-check once it exists
Inferara/inference#499 · opened on Sep 27, 2026
Measure whether SpaceWasm builds need their own unroll thresholds
Inferara/inference#498 · opened on Sep 27, 2026
Revisit WASI imports under infs run
Inferara/inference#478 · opened on Sep 23, 2026

Navigation Menu

Repository navigation

Issues

Issues

CHANGELOG: the #221 Breaking entry names `ManifestSettings` fields that do not exist

infs: the `memory:` build line shows the requested stack, not the one static data leaves

infs: re-check a defined memory's limits after `[build.wasm-opt]`

SpaceWasm builds with an imported linear memory

Link `[wasm-dependencies]` modules into a module that imports its memory

Module-scope constants (#515): follow-up tracker

Module constants: tests for guarantees nothing currently enforces

Module constants: documentation, diagnostic wording and CHANGELOG corrections

A module constant whose type is item-imported fails to type-check

Module constants in specifications: constant indices aren't folded, and tables over 64 elements can't be indexed symbolically

A036 and A058 suggest memory layouts that fail the other check or that the build refuses

Module constants: place static data per build mode and hand analysis and codegen one placed layout

Compute type layout once: codegen, A036 and the constant evaluator each have their own

Module constants: initializers skip most analysis rules, and constness depends on values

Module constants: the evaluator panics, aborts or emits an invalid module on user input

wasm-to-v: name-handling follow-ups from #405 (stdlib import capture, embedded module name, spec-name over-rejection, stale hints and docs)

`continue;` is reported as an undeclared variable, and compiles as a no-op when a binding named `continue` is in scope

Stellar: `contractspecv0` has an empty doc for every method; `///` doc comments never reach it

book: Stellar proving procedure says `proofs/contract.wasm` is the pre-rewrite module byte for byte, but a `spec` block makes it a superset

infs has no proof build for a `stellar` or `spacewasm` project without editing `[build] target`, and the refused proof build deletes the previous artifact

Overflow guards cost the ray-tracer benchmark 72% of its throughput

Allow bulk memory for SpaceWasm builds once nasa/spacewasm#54 ships

Check the SpaceWasm limits against upstream `spacewasm-check` once it exists

Measure whether SpaceWasm builds need their own unroll thresholds

Revisit WASI imports under `infs run`