d82ebcc102
* x64: Enable load-coalescing for SSE/AVX instructions This commit unlocks the ability to fold loads into operands of SSE and AVX instructions. This is beneficial for both function size when it happens in addition to being able to reduce register pressure. Previously this was not done because most SSE instructions require memory to be aligned. AVX instructions, however, do not have alignment requirements. The solution implemented here is one recommended by Chris which is to add a new `XmmMemAligned` newtype wrapper around `XmmMem`. All SSE instructions are now annotated as requiring an `XmmMemAligned` operand except for a new new instruction styles used specifically for instructions that don't require alignment (e.g. `movdqu`, `*sd`, and `*ss` instructions). All existing instruction helpers continue to take `XmmMem`, however. This way if an AVX lowering is chosen it can be used as-is. If an SSE lowering is chosen, however, then an automatic conversion from `XmmMem` to `XmmMemAligned` kicks in. This automatic conversion only fails for unaligned addresses in which case a load instruction is emitted and the operand becomes a temporary register instead. A number of prior `Xmm` arguments have now been converted to `XmmMem` as well. One change from this commit is that loading an unaligned operand for an SSE instruction previously would use the "correct type" of load, e.g. `movups` for f32x4 or `movup` for f64x2, but now the loading happens in a context without type information so the `movdqu` instruction is generated. According to [this stack overflow question][question] it looks like modern processors won't penalize this "wrong" choice of type when the operand is then used for f32 or f64 oriented instructions. Finally this commit improves some reuse of logic in the `put_in_*_mem*` helper to share code with `sinkable_load` and avoid duplication. With this in place some various ISLE rules have been updated as well. In the tests it can be seen that AVX-instructions are now automatically load-coalesced and use memory operands in a few cases. [question]: https://stackoverflow.com/questions/40854819/is-there-any-situation-where-using-movdqu-and-movupd-is-better-than-movups * Fix tests * Fix move-and-extend to be unaligned These don't have alignment requirements like other xmm instructions as well. Additionally add some ISA tests to ensure that their output is tested. * Review comments
Cranelift Code Generator
A Bytecode Alliance project
Cranelift is a low-level retargetable code generator. It translates a target-independent intermediate representation into executable machine code.
For more information, see the documentation.
For an example of how to use the JIT, see the JIT Demo, which implements a toy language.
For an example of how to use Cranelift to run WebAssembly code, see Wasmtime, which implements a standalone, embeddable, VM using Cranelift.
Status
Cranelift currently supports enough functionality to run a wide variety of programs, including all the functionality needed to execute WebAssembly (MVP and various extensions like SIMD), although it needs to be used within an external WebAssembly embedding such as Wasmtime to be part of a complete WebAssembly implementation. It is also usable as a backend for non-WebAssembly use cases: for example, there is an effort to build a Rust compiler backend using Cranelift.
Cranelift is production-ready, and is used in production in several places, all within the context of Wasmtime. It is carefully fuzzed as part of Wasmtime with differential comparison against V8 and the executable Wasm spec, and the register allocator is separately fuzzed with symbolic verification. There is an active effort to formally verify Cranelift's instruction-selection backends. We take security seriously and have a security policy as a part of Bytecode Alliance.
Cranelift has three backends: x86-64, aarch64 (aka ARM64), and s390x (aka IBM Z). All three backends fully support enough functionality for Wasm MVP, and x86-64 and aarch64 fully support SIMD as well. On x86-64, Cranelift supports both the System V AMD64 ABI calling convention used on many platforms and the Windows x64 calling convention. On aarch64, Cranelift supports the standard Linux calling convention and also has specific support for macOS (i.e., M1 / Apple Silicon).
Cranelift's code quality is within range of competitiveness to browser JIT engines' optimizing tiers. A recent paper includes third-party benchmarks of Cranelift, driven by Wasmtime, against V8 and an LLVM-based Wasm engine, WAVM (Fig 22). The speed of Cranelift's generated code is ~2% slower than that of V8 (TurboFan), and ~14% slower than WAVM (LLVM). Its compilation speed, in the same paper, is measured as approximately an order of magnitude faster than WAVM (LLVM). We continue to work to improve both measures.
The core codegen crates have minimal dependencies and are carefully written to handle malicious or arbitrary compiler input: in particular, they do not use callstack recursion.
Cranelift performs some basic mitigations for Spectre attacks on heap bounds checks, table bounds checks, and indirect branch bounds checks; see #1032 for more.
Cranelift's APIs are not yet considered stable, though we do follow semantic-versioning (semver) with minor-version patch releases.
Cranelift generally requires the latest stable Rust to build as a policy, and is tested as such, but we can incorporate fixes for compilation with older Rust versions on a best-effort basis.
Contributing
If you're interested in contributing to Cranelift: thank you! We have a contributing guide which will help you getting involved in the Cranelift project.
Planned uses
Cranelift is designed to be a code generator for WebAssembly, but it is general enough to be useful elsewhere too. The initial planned uses that affected its design were:
- Wasmtime non-Web wasm engine.
- Debug build backend for the Rust compiler.
- WebAssembly compiler for the SpiderMonkey engine in Firefox (currently not planned anymore; SpiderMonkey team may re-assess in the future).
- Backend for the IonMonkey JavaScript JIT compiler in Firefox (currently not planned anymore; SpiderMonkey team may re-assess in the future).
Building Cranelift
Cranelift uses a conventional Cargo build process.
Cranelift consists of a collection of crates, and uses a Cargo
Workspace,
so for some cargo commands, such as cargo test, the --all is needed
to tell cargo to visit all of the crates.
test-all.sh at the top level is a script which runs all the cargo
tests and also performs code format, lint, and documentation checks.
Log configuration
Cranelift uses the log crate to log messages at various levels. It doesn't
specify any maximal logging level, so embedders can choose what it should be;
however, this can have an impact of Cranelift's code size. You can use log
features to reduce the maximum logging level. For instance if you want to limit
the level of logging to warn messages and above in release mode:
[dependency.log]
...
features = ["release_max_level_warn"]
Editor Support
Editor support for working with Cranelift IR (clif) files: