From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: qemu-rust@nongnu.org, zhao1.liu@intel.com
Subject: [PATCH v2 00/12] rust: wrap all C types exposed through qemu_api
Date: Thu, 27 Feb 2025 15:22:07 +0100 [thread overview]
Message-ID: <20250227142219.812270-1-pbonzini@redhat.com> (raw)
This is the second part of "rust: prepare for splitting crates" with
Zhao's suggestions addressed and with more precise handling of pinning
for Timers.
The series introduce a third generic type in qemu_api::cell, Opaque<T>.
This type is similar to a same-named type in Linux; it is basically a
"disable all undefined behavior" switch for the Rust compiler and it
helps maintaining safety at the Rust/C boundary, complementing the
existing BqlCell and BqlRefCell types.
Apart from making things more formally correct, this makes it possible
to implement methods on a struct that is distinct from the one produced
by bindgen. This has a couple of advantages:
- you do not have to disable the Copy trait on structs where you want
to add a Drop trait. This was already a problem for the Timer struct.
- whether Send and Sync are implemented is entirely a decision of the
place that implements the wrapper. Previously, a struct with no
pointers for example would have been always both Send and Sync,
whereas now that can be adjusted depending on the actual
thread-safety of the Rust methods.
- more pertinent to the "multiple crates" plan that prompted posting
of v1, you do not have to put the methods in the same crate as the
bindgen-generated bindings.inc.rs.
It also makes Debug output a bit less unwieldy, and in the future one
might want to add specialized implementations of Display and Debug that
are both useful and readable.
Paolo
Supersedes: <20250221170342.63591-1-pbonzini@redhat.com>
v1->v2:
From Zhao's review:
- fix Opaque::zeroed()
- improve comments for as_mut_ptr() fand as_void_ptr()
- remove unnecessary ".0" accesses, or highlight why they're needed
- add patch to access SysBusDevice MMIO addresses safely
Other changes:
- improve safety comments for constructors of Opaque
- remove implementation of Default for Opaque<T: Default>,
leaving Opaque::new() in but only as an unsafe function
- change Timer patch to construct timers as Pin<Box<Self>>,
following the documentation of `Opaque<>`
- add "rust: vmstate: add std::pin::Pin as transparent wrapper"
Paolo Bonzini (12):
rust: cell: add wrapper for FFI types
rust: qemu_api_macros: add Wrapper derive macro
rust: vmstate: add std::pin::Pin as transparent wrapper
rust: timer: wrap QEMUTimer with Opaque<> and express pinning
requirements
rust: irq: wrap IRQState with Opaque<>
rust: qom: wrap Object with Opaque<>
rust: qdev: wrap Clock and DeviceState with Opaque<>
rust: hpet: do not access fields of SysBusDevice
rust: sysbus: wrap SysBusDevice with Opaque<>
rust: memory: wrap MemoryRegion with Opaque<>
rust: chardev: wrap Chardev with Opaque<>
rust: bindings: remove more unnecessary Send/Sync impls
docs/devel/rust.rst | 36 +++--
meson.build | 7 -
rust/hw/timer/hpet/src/hpet.rs | 27 ++--
rust/qemu-api-macros/src/lib.rs | 86 +++++++++++-
rust/qemu-api/meson.build | 7 +-
rust/qemu-api/src/bindings.rs | 26 +---
rust/qemu-api/src/cell.rs | 228 +++++++++++++++++++++++++++++++-
rust/qemu-api/src/chardev.rs | 8 +-
rust/qemu-api/src/irq.rs | 15 ++-
rust/qemu-api/src/memory.rs | 37 +++---
rust/qemu-api/src/qdev.rs | 75 +++++++----
rust/qemu-api/src/qom.rs | 35 +++--
rust/qemu-api/src/sysbus.rs | 40 +++++-
rust/qemu-api/src/timer.rs | 47 ++++---
rust/qemu-api/src/vmstate.rs | 3 +-
15 files changed, 527 insertions(+), 150 deletions(-)
--
2.48.1
next reply other threads:[~2025-02-27 14:22 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-02-27 14:22 Paolo Bonzini [this message]
2025-02-27 14:22 ` [PATCH 01/12] rust: cell: add wrapper for FFI types Paolo Bonzini
2025-02-27 14:22 ` [PATCH 02/12] rust: qemu_api_macros: add Wrapper derive macro Paolo Bonzini
2025-02-27 14:22 ` [PATCH 03/12] rust: vmstate: add std::pin::Pin as transparent wrapper Paolo Bonzini
2025-03-03 13:25 ` Zhao Liu
2025-02-27 14:22 ` [PATCH 04/12] rust: timer: wrap QEMUTimer with Opaque<> and express pinning requirements Paolo Bonzini
2025-03-03 13:48 ` Zhao Liu
2025-03-03 15:58 ` Paolo Bonzini
2025-03-04 9:13 ` Zhao Liu
2025-03-06 10:45 ` Paolo Bonzini
2025-03-06 11:35 ` Zhao Liu
2025-03-03 14:28 ` Zhao Liu
2025-03-03 14:51 ` Paolo Bonzini
2025-03-03 16:15 ` Zhao Liu
2025-02-27 14:22 ` [PATCH 05/12] rust: irq: wrap IRQState with Opaque<> Paolo Bonzini
2025-03-03 15:07 ` Zhao Liu
2025-02-27 14:22 ` [PATCH 06/12] rust: qom: wrap Object " Paolo Bonzini
2025-02-27 14:22 ` [PATCH 07/12] rust: qdev: wrap Clock and DeviceState " Paolo Bonzini
2025-02-27 14:22 ` [PATCH 08/12] rust: hpet: do not access fields of SysBusDevice Paolo Bonzini
2025-03-03 15:09 ` Zhao Liu
2025-02-27 14:22 ` [PATCH 09/12] rust: sysbus: wrap SysBusDevice with Opaque<> Paolo Bonzini
2025-03-03 15:19 ` Zhao Liu
2025-02-27 14:22 ` [PATCH 10/12] rust: memory: wrap MemoryRegion " Paolo Bonzini
2025-03-03 15:25 ` Zhao Liu
2025-03-05 7:09 ` Zhao Liu
2025-02-27 14:22 ` [PATCH 11/12] rust: chardev: wrap Chardev " Paolo Bonzini
2025-02-27 14:22 ` [PATCH 12/12] rust: bindings: remove more unnecessary Send/Sync impls Paolo Bonzini
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20250227142219.812270-1-pbonzini@redhat.com \
--to=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=qemu-rust@nongnu.org \
--cc=zhao1.liu@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).