[46.0.0] Some more backports#13626
Open
alexcrichton wants to merge 8 commits into
Open
Conversation
…tecodealliance#13608) * add test for `substituted_component_type` panic `Linker::substituted_component_type` builds a `types::Component` whose resource substitution map is `Some(..)` but only covers the component's *imported* resources. Introspecting an exported function that references an *exported* (non-imported) resource then panics with an index-out-of-bounds, because `InstanceType::resource_type` hard-indexes that partial map (`matching.rs`) instead of falling back to treating an absent resource as uninstantiated. This adds a test that builds a component exporting a resource plus a constructor returning `own<t>` and a method taking `borrow<t>`, then walks the introspected function types via `substituted_component_type`. It panics today (index out of bounds) and will pass once the resolver falls back gracefully. Assisted-by: claude:claude-opus-4-8 * fix panic on guest-exported resources `Linker::substituted_component_type` constructs the introspection type with `resources: Some(&cx.imported_resources)`, a map that only contains the component's *imported* resources. `InstanceType::resource_type` previously hard-indexed that map (`self.resources.map(|t| t[ty])`), so resolving a resource that is not in the map -- e.g. a resource the component *exports* -- panicked with an index-out-of-bounds whenever an exported function's params/results referenced it. Fix this by using a fallible lookup and falling back to `ResourceType::uninstantiated` when the resource is absent from the map, exactly as the `resources: None` path already does. This is the semantically correct behavior: a resource that has no substitution is "not (yet) instantiated", which is also what `Component::component_type()` relies on (it passes `resources: None`). With this fix the same safe, public introspection API no longer panics depending on which (also public) API produced the `types::Component`. Assisted-by: claude:claude-opus-4-8
Integration upstream into rust-lang/rust will be a bit easier with this relaxed bound, so widen the version requirement here a bit.
The number of failures right now is... high. Bake a `cargo fetch` directly into all Rust installations on CI and put a loop around it.
Throw out timeouts and try to handle them more aggressively to see if this can reduce the number of flaky failures we see in CI...
…13612) * Update WASI to 0.3.0, enable component-model-async This commit updates the vendored WASI WITs for the 0.3.0 release of WASI to this repository, updating the supported version of WASI that `wasmtime-wasi` runs. This additionally enables the `component-model-async` wasm feature by default in Wasmtime, along with `-Sp3` on the CLI as well. This is intended to be a comprehensive "turn WASI 0.3.0 on by default" PR which touches a few different locations. Apart from changing version numbers some minor changes made here are: * Default enablement of the wasm `cm-async` feature is now conditional on `Config::concurrency_support` in addition to the compile-time feature. This means that if `concurrency_support` is disabled the default will be that `cm-async` is disabled. * The `tests/wasi.rs` test suite running the upstream `wasi-testsuite` repository is updated to expect failures for all WASIp3 tests due to the import versions changing. This resulted in a necessary restructuring of the test to handle a few more failures in a few more locations to ensure that "should fail tests" are correctly marked as passing when they indeed do fail. * Fix handling an erroneous exit * Review comments
alexcrichton
requested review from
fitzgen and
pchickey
and removed request for
a team
* pulley: add regression test for clobber-save slot collision A `preserve_all` frame on pulley can save both manually-managed clobbers (x0-x15, floats, vectors) and pulley-managed clobbers (x16+, saved at the top of the frame by `push_frame_save`). When both are present, a manually-managed register could be assigned the same stack slot that `push_frame_save` already uses for a pulley-managed register, so the two stores collide and corrupt the saved value. Add a `preserve-all.clif` case (pulley32 and pulley64) whose frame holds a value live across a call in a pulley-managed callee-saved register (x16) while the calls clobber x0-x15. The committed precise output is the correct layout (x0 saved below the pulley-saved region); without the layout fix the emitted code stores both x16 and x0 to the same offset. * pulley: fix clobber-save stack-slot collision `FrameLayout::manually_managed_clobbers` walked `clobbered_callee_saves` from the top of the frame assuming the pulley-managed registers (saved by `push_frame_save`) come first. They do not: `clobbered_callee_saves` is sorted by register, and the manually-managed integer registers x0-x15 sort before the pulley-managed x16-x31. As a result the first manually-managed register was assigned the topmost slot, which `push_frame_save` already uses for the first pulley-saved register, so the two stores collided and the pulley-saved register was restored from a clobbered slot. Reserve the top `num_saved_by_pulley` 8-byte slots for the pulley-managed registers (matching `push_frame_save`) and hand the lower slots to the manually-managed registers, eliminating the overlap.
Because apparently we need retry loops on everything now
github-actions
Bot
added
cranelift
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Bundling a number of PRs for the 46.0.0 release branch:
cc#13617cargo fetchin a retry loop on CI #13624