RustInQemu: Difference between revisions

From QEMU
Line 55: Line 55:
* QOM (https://gitlab.com/bonzini/qemu/-/commits/rust-qom (defining QOM classes entirely without macros, QOM casts and reference counting)
* QOM (https://gitlab.com/bonzini/qemu/-/commits/rust-qom (defining QOM classes entirely without macros, QOM casts and reference counting)
** also adds documentation for public APIs in qemu-api and qemu-api-macros
** also adds documentation for public APIs in qemu-api and qemu-api-macros
* automatically generate rustc flags from Cargo.toml
* Better integration with Cargo subcommands
* Better integration with Cargo subcommands
** Perform clippy checks ("ninja clippy"?
** automatically generate rustc flags from Cargo.toml
** Reformat code ("ninja rustfmt"?
** Perform clippy checks and reformat code ("ninja clippy", "ninja rustfmt") - TODO: rustfmt configuration currently requires nightly due to use of unstable features
* More CI
* More CI
** Rustfmt checks
** Rustfmt checks

Revision as of 19:02, 31 October 2024

For the old RustInQemu page, see RustInQemu/2022

Active efforts in 2024

  • Subject: [RFC 0/6] scripts: Rewrite simpletrace printer in Rust
    Date: Mon, 27 May 2024 16:14:15 +0800
    RFC v1
  • ARM PL011 UART device model in Rust
    Subject: [RFC PATCH v1 0/6] Implement ARM PL011 in Rust
    Date: Mon, 10 Jun 2024 21:22:35 +0300
    RFC v1 v6
    • Meson and configure integration
    • Bindings generation

After some discussion we decided to use rustc directly without going through cargo and each external crate becomes a Meson subproject. Meson 1.5.0+'s support for method = cargo is pretty good for host-compiled packages, but it doesn't work for dependencies of build-compiled packages (i.e. dependencies of procedural macros). This is tracked by https://github.com/mesonbuild/meson/issues/11121. Until then, meson.build files for procedural macros and their dependencies have to be written by hand. (Currently _all_ dependencies use hand-written meson.build).

While cargo would be fine as an interim solution, using Meson to compile the Rust code removes some moving parts. The extra cost of writing meson.build files for dependencies is expected to be small and will mostly happen upfront.

Past efforts

  • [RFC v3 00/32] Rust binding for QAPI and qemu-ga QMP handler examples
    on patchew on lore

Minimum supported versions

1.59.0 has const CStr::from_bytes_with_nul_unchecked, not really possible to go below that.

Bindgen 0.60.x has --allowlist-file. Ubuntu 22.04 has 0.59.x which is a problem.

In the future 1.83.0 would allow nested offset_of.

Done

  • only generate bindings.rs.inc once
  • disabling the cast_ptr_alignment lint is too broad
  • update bundled meson to 1.5.0
  • add support for --rustc and RUSTC environment variables
  • remove --no-include-path-detection from bindgen invocation
  • rename subprojects to use Meson 1.5.0+ convention and add meson.override_dependency() call
  • add support for cross compilation of Rust subprojects (native: true for deps of procedural macro crates)
  • fix licenses for Rust subprojects

Posted

  • Enable use of rustc 1.63.0
    • fix cfgs for proc-macro2
    • core::ffi (stable in 1.64.0)
    • "if let" usage in bilge-impl needs to be patched out (stable in 1.65.0)
    • std::sync::OnceLock (stable in 1.70.0)
    • MaybeUninit::zeroed() (stable in 1.75.0 in const context)
    • c"" literals (stable in 1.77.0)
    • offset_of! (stable in 1.77.0)
  • Enable use of older bindgen
    • What to do about Ubuntu 22.04? Bindgen 0.59.x does not have --allowlist-file
  • Install rustc/bindgen in all containers

To be posted

TODO

  • Remove need for manual "meson subprojects update --reset" when updating packagefiles/
  • Improve cargo integration
    • Run code checks a la "meson test --suite codecheck"? (clippy and rustfmt)
    • Documentation generation ("ninja rustdoc"? Maybe check how Linux runs doctests?)
  • rustfmt currently requires nightly, decide what to do about it
  • Place rustdoc output for master somewhere?
  • feature parity for pl011
    • check for new commits
  • bindings
    • MemoryRegion (callbacks)
    • Chardev (callbacks, functions)
    • DMA (investigate vm-memory?)
  • other experiments: https://github.com/bonzini/rust-qemu (Generic Rust<->C interop, Error)
  • more QOM procedural macros (currently #[derive(Object)])
    • generate qdev properties?
    • generate parts of TypeInfo?

Ideas for lints without breaking CI

See https://github.com/bonzini/rust-qemu/commit/95b25f7c5f4e2694a85a5503050cc98da7562c7c

  • run clippy as part of "make check", possibly only if Rust is newer than some version (1.74.0 so that clippy can be configured in Cargo.toml?)
  • deny many individual lints, do not deny groups (complexity, perf, style, suspicious) on regular builds. allow unknown_lints.
  • add to CI a fallible job that runs on nightly clippy with -Dclippy::complexity -Dclippy::perf -Dclippy::suspicious -Dclippy::style -Dunknown_lints. the job should generally pass, and if a new lint triggers it probably should be added to Cargo.toml as either "allow" (rare) or "deny" (possibly after adding #[allow()] to the source).

Possible project targets

Miscellanea

- qemu-bridge-helper.c Re-write SUID C executable with useful features.

Devices

  • hw/block/pflash_cfi01.c, hw/block/pflash_cfi02.c (claimed, WIP)
    • needs block bindings
  • hw/timer/i8254.c (claimed, WIP)
    • needs timer bindgings
  • hw/mem/nvdimm.c (suggested by Manos)