rust-for-linux.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: Lyude Paul <lyude@redhat.com>, rust-for-linux@vger.kernel.org
Cc: "Danilo Krummrich" <dakr@redhat.com>,
	airlied@redhat.com, "Ingo Molnar" <mingo@redhat.com>,
	"Will Deacon" <will@kernel.org>,
	"Waiman Long" <longman@redhat.com>,
	"Peter Zijlstra" <peterz@infradead.org>,
	linux-kernel@vger.kernel.org,
	"Benno Lossin" <benno.lossin@proton.me>,
	"Daniel Almeida" <daniel.almeida@collabora.com>,
	"Gary Guo" <gary@garyguo.net>, "Miguel Ojeda" <ojeda@kernel.org>,
	"Alex Gaynor" <alex.gaynor@gmail.com>,
	"Wedson Almeida Filho" <wedsonaf@gmail.com>,
	"Boqun Feng" <boqun.feng@gmail.com>,
	"Björn Roy Baron" <bjorn3_gh@protonmail.com>,
	"Andreas Hindborg" <a.hindborg@samsung.com>,
	"Alice Ryhl" <aliceryhl@google.com>,
	"Trevor Gross" <tmgross@umich.edu>,
	"Martin Rodriguez Reboredo" <yakoyoku@gmail.com>,
	"Valentin Obst" <kernel@valentinobst.de>
Subject: Re: [PATCH v6 3/3] rust: sync: Add SpinLockIrq
Date: Wed, 02 Oct 2024 22:53:32 +0200	[thread overview]
Message-ID: <8734lew7jn.ffs@tglx> (raw)
In-Reply-To: <20240916213025.477225-4-lyude@redhat.com>

On Mon, Sep 16 2024 at 17:28, Lyude Paul wrote:
> A variant of SpinLock that is expected to be used in noirq contexts, and
> thus requires that the user provide an kernel::irq::IrqDisabled to prove
> they are in such a context upon lock acquisition. This is the rust
> equivalent of spin_lock_irqsave()/spin_lock_irqrestore().

This fundamentally does not work with CONFIG_PREEMPT_RT. See:

   https://www.kernel.org/doc/html/latest/locking/locktypes.html

for further information. TLDR:

On RT enabled kernels spin/rw_lock are substituted by sleeping locks. So
you _cannot_ disable interrupts before taking the lock on RT enabled
kernels. That will result in a 'might_sleep()' splat.

There is a reason why the kernel has two distinct spinlock types:

    raw_spinlock_t and spinlock_t

raw_spinlock_t is a real spinning lock independent of CONFIG_PREEMPT_RT,
spinlock_t is mapped to raw_spinlock_t on CONFIG_PREEMPT_RT=n and to a
rtmutex based implementation for CONFIG_PREEMPT_RT=y.

As a consequence spin_lock_irq() and spin_lock_irqsave() will _NOT_
disable interrupts on a CONFIG_PREEMPT_RT=y kernel.

The proposed rust abstraction is not suitable for that.

At this phase of rust integration there is no need to wrap
raw_spinlock_t, so you have two options to solve that:

   1) Map Rust's SpinLockIrq() to spin_lock_irqsave() and
      spin_unlock_irqrestore() which does the right thing

   2) Play all the PREEMPT_RT games in the local irq disable abstraction

#1 is the right thing to do because no driver should rely on actually
disabling interrupts on the CPU. If there is a driver which does that,
then it's not compatible with RT and should use a local lock instead.

local locks aside of being RT compatible have the benefit that they give
scope to the protected region/data, while a plain local_irq_disable()
does not.

Don't even think about exposing this 'with_irq_disabled' interface
unless you are trying to move actual core code like the scheduler or low
level interrupt handling to rust.

Create explicit interrupt safe interfaces which map to the underlying
locking primitives instead.

Thanks,

        tglx



  parent reply	other threads:[~2024-10-02 20:53 UTC|newest]

Thread overview: 75+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-09-16 21:28 [PATCH v6 0/3] rust: Add irq abstraction, SpinLockIrq Lyude Paul
2024-09-16 21:28 ` [PATCH v6 1/3] rust: Introduce irq module Lyude Paul
2024-09-29 20:36   ` Trevor Gross
2024-09-29 23:45   ` Boqun Feng
2024-10-02 20:20   ` Thomas Gleixner
2024-10-04  8:58     ` Benno Lossin
2024-10-04 17:18       ` Lyude Paul
2024-10-17 18:51       ` Lyude Paul
2024-10-04 17:02     ` Lyude Paul
2024-10-10 21:00   ` Daniel Almeida
2024-09-16 21:28 ` [PATCH v6 2/3] rust: sync: Introduce lock::Backend::Context Lyude Paul
2024-09-29 20:40   ` Trevor Gross
2024-09-29 23:52   ` Boqun Feng
2024-09-16 21:28 ` [PATCH v6 3/3] rust: sync: Add SpinLockIrq Lyude Paul
2024-09-29 20:50   ` Trevor Gross
2024-09-29 23:59   ` Boqun Feng
2024-10-02 20:53   ` Thomas Gleixner [this message]
2024-10-03 12:51     ` Boqun Feng
2024-10-04 18:48     ` Lyude Paul
2024-10-05 18:19       ` Lyude Paul
2024-10-07 12:42         ` Boqun Feng
2024-10-07 18:13           ` Lyude Paul
2024-10-15 12:57           ` Andreas Hindborg
2024-10-15 20:17             ` Boqun Feng
2024-10-15 20:21               ` Boqun Feng
2024-10-16 20:57                 ` Lyude Paul
2024-10-17 13:34                   ` Andreas Hindborg
2024-10-07 12:01       ` Thomas Gleixner
2024-10-07 18:30         ` Lyude Paul
2024-10-08 15:21           ` Thomas Gleixner
2024-10-12  8:01             ` Boqun Feng
2024-10-10 16:39 ` [PATCH v6 0/3] rust: Add irq abstraction, SpinLockIrq Daniel Almeida
2024-10-12  5:29 ` Dirk Behme
2024-10-13 19:06   ` Thomas Gleixner
2024-10-13 21:43     ` Boqun Feng
2024-10-16 21:00       ` Thomas Gleixner
2024-10-16 21:31         ` Boqun Feng
2024-10-17 20:49           ` Lyude Paul
2024-10-17 22:27             ` Boqun Feng
2024-10-18  5:51           ` [POC 0/6] Allow SpinLockIrq to use a normal Guard interface Boqun Feng
2024-10-18  5:51             ` [POC 1/6] irq & spin_lock: Add counted interrupt disabling/enabling Boqun Feng
2024-10-21  7:04               ` kernel test robot
2024-10-21  7:35               ` kernel test robot
2024-10-21 20:44               ` Lyude Paul
2024-10-24 16:18                 ` Peter Zijlstra
2024-10-23 19:34               ` Thomas Gleixner
2024-10-23 19:51                 ` Peter Zijlstra
2024-10-23 20:38                   ` Thomas Gleixner
2024-10-24 10:05                     ` Peter Zijlstra
2024-10-24 17:22                       ` Thomas Gleixner
2024-10-24 21:57                         ` Boqun Feng
2024-10-25 15:04                           ` Thomas Gleixner
2024-10-25 18:28                             ` Peter Zijlstra
2024-10-24 19:12                       ` Lyude Paul
2025-07-24 20:36                   ` w/r/t "irq & spin_lock: Add counted interrupt disabling/enabling": holes in pcpu_hot? Lyude Paul
2025-07-24 21:59                     ` Thomas Gleixner
2024-10-24  5:05                 ` [POC 1/6] irq & spin_lock: Add counted interrupt disabling/enabling Boqun Feng
2024-10-24  8:17                   ` Thomas Gleixner
2024-10-24 16:20                     ` Boqun Feng
2024-10-18  5:51             ` [POC 2/6] rust: Introduce interrupt module Boqun Feng
2024-10-31 20:45               ` Lyude Paul
2024-10-31 20:47                 ` Lyude Paul
2024-10-18  5:51             ` [POC 3/6] rust: helper: Add spin_{un,}lock_irq_{enable,disable}() helpers Boqun Feng
2024-10-18  5:51             ` [POC 4/6] rust: sync: Add SpinLockIrq Boqun Feng
2024-10-18 19:23               ` Lyude Paul
2024-10-18 20:22                 ` Boqun Feng
2024-10-18  5:51             ` [POC 5/6] rust: sync: Introduce lock::Backend::Context Boqun Feng
2024-10-31 20:54               ` Lyude Paul
2024-10-18  5:51             ` [POC 6/6] rust: sync: lock: Add `Backend::BackendInContext` Boqun Feng
2024-10-18 10:22             ` [POC 0/6] Allow SpinLockIrq to use a normal Guard interface Andreas Hindborg
2024-10-18 12:42               ` Boqun Feng
2024-10-18 11:16             ` Andreas Hindborg
2024-10-18 16:05               ` Dirk Behme
2024-10-31 20:56             ` Lyude Paul
2024-10-17 20:42         ` [PATCH v6 0/3] rust: Add irq abstraction, SpinLockIrq Lyude Paul

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=8734lew7jn.ffs@tglx \
    --to=tglx@linutronix.de \
    --cc=a.hindborg@samsung.com \
    --cc=airlied@redhat.com \
    --cc=alex.gaynor@gmail.com \
    --cc=aliceryhl@google.com \
    --cc=benno.lossin@proton.me \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun.feng@gmail.com \
    --cc=dakr@redhat.com \
    --cc=daniel.almeida@collabora.com \
    --cc=gary@garyguo.net \
    --cc=kernel@valentinobst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=longman@redhat.com \
    --cc=lyude@redhat.com \
    --cc=mingo@redhat.com \
    --cc=ojeda@kernel.org \
    --cc=peterz@infradead.org \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=tmgross@umich.edu \
    --cc=wedsonaf@gmail.com \
    --cc=will@kernel.org \
    --cc=yakoyoku@gmail.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).