From: Shrikanth Hegde <sshegde@linux.ibm.com>
To: Boqun Feng <boqun@kernel.org>, Peter Zijlstra <peterz@infradead.org>
Cc: "Ingo Molnar" <mingo@kernel.org>, "Will Deacon" <will@kernel.org>,
"Waiman Long" <longman@redhat.com>, "Gary Guo" <gary@garyguo.net>,
"Alice Ryhl" <aliceryhl@google.com>,
"Lyude Paul" <lyude@redhat.com>,
"Daniel Almeida" <daniel.almeida@collabora.com>,
"Onur Özkan" <work@onurozkan.dev>,
"Miguel Ojeda" <ojeda@kernel.org>,
"Danilo Krummrich" <dakr@kernel.org>,
linux-kernel@vger.kernel.org, rust-for-linux@vger.kernel.org
Subject: Re: [PATCH v4 05/17] irq & spin_lock: Add counted interrupt disabling/enabling
Date: Wed, 5 Aug 2026 02:21:49 +0530 [thread overview]
Message-ID: <2fc01d90-e081-4ddf-a842-b68eda15ca9e@linux.ibm.com> (raw)
In-Reply-To: <20260804161447.84806-6-boqun@kernel.org>
On 8/4/26 9:44 PM, Boqun Feng wrote:
> Currently the nested interrupt disabling and enabling is represented by
> _irqsave() and _irqrestore() APIs, which are relatively unsafe, for
> example:
>
> <interrupts are enabled as beginning>
> spin_lock_irqsave(l1, flag1);
> spin_lock_irqsave(l2, flag2);
> spin_unlock_irqrestore(l1, flags1);
> <l2 is still held but interrupts are enabled>
> // accesses to interrupt-disable protected data will cause races
>
> This is even easier to trigger with guard facilities:
>
> unsigned long flag2;
>
> scoped_guard(spin_lock_irqsave, l1) {
> spin_lock_irqsave(l2, flag2);
> }
> // l2 locked but interrupts are enabled.
> spin_unlock_irqrestore(l2, flag2);
>
> (Hand-to-hand locking critical sections are not uncommon for a
> fine-grained lock design)
>
> And because of this unsafety, Rust cannot easily wrap the
> interrupt-disabling locks in a safe API, which complicates the design.
>
> To resolve this, introduce a new set of interrupt disabling APIs:
>
> * local_interrupt_disable();
> * local_interrupt_enable();
>
> They work like local_irq_save() and local_irq_restore() except that 1)
> the outermost local_interrupt_disable() call saves the interrupt state
> into a per-CPU variable, so that the outermost local_interrupt_enable()
> can restore the state, and 2) a per-CPU counter is added to record the
> nest level of these calls, so that interrupts are not accidentally
> enabled inside the outermost critical section.
>
> Also add the corresponding spin_lock primitives: spin_lock_irq_disable()
> and spin_unlock_irq_enable(), as a result, code as follows:
>
> spin_lock_irq_disable(l1);
> spin_lock_irq_disable(l2);
> spin_unlock_irq_enable(l1);
> // Interrupts are still disabled.
> spin_unlock_irq_enable(l2);
>
> doesn't have the issue that interrupts are accidentally enabled.
>
> This also makes the wrapper of interrupt-disabling locks on Rust easier
> to design.
>
> Signed-off-by: Lyude Paul <lyude@redhat.com>
> [boqun: Apply Peter's feedback and fix spell errors reported by Ingo]
> Signed-off-by: Boqun Feng <boqun@kernel.org>
> ---
> include/linux/interrupt_rc.h | 82 ++++++++++++++++++++++++++++++++
> include/linux/preempt.h | 4 ++
> include/linux/spinlock.h | 23 +++++++++
> include/linux/spinlock_api_smp.h | 43 +++++++++++++++++
> include/linux/spinlock_api_up.h | 15 ++++++
> include/linux/spinlock_rt.h | 18 +++++++
> kernel/locking/spinlock.c | 31 ++++++++++++
> kernel/softirq.c | 28 ++++++++++-
> 8 files changed, 242 insertions(+), 2 deletions(-)
> create mode 100644 include/linux/interrupt_rc.h
>
> diff --git a/include/linux/interrupt_rc.h b/include/linux/interrupt_rc.h
> new file mode 100644
> index 000000000000..b9a7f05ecf42
> --- /dev/null
> +++ b/include/linux/interrupt_rc.h
> @@ -0,0 +1,82 @@
> +/* SPDX-License-Identifier: GPL-2.0 */
> +#ifndef __LINUX_INTERRUPT_RC_H
> +#define __LINUX_INTERRUPT_RC_H
> +
> +/*
> + * include/linux/interrupt_rc.h - refcounted local processor interrupt
> + * management.
> + *
> + * Since the implementation of this API currently depends on
> + * local_irq_save()/local_irq_restore(), we split this into its own header to
> + * make it easier to include without hitting circular header dependencies.
> + */
> +
> +#include <linux/irqflags.h>
> +#include <linux/preempt.h>
> +#include <linux/processor.h>
> +#include <linux/smp.h>
> +
> +#ifndef MODULE
> +/* Per-CPU interrupt disabling state for local_interrupt_{disable,enable}(). */
> +DECLARE_PER_CPU(unsigned long, local_interrupt_disable_state);
> +
> +static __always_inline void __local_interrupt_disable(void)
> +{
> + unsigned long flags;
> +
> + local_irq_save(flags);
> + raw_cpu_write(local_interrupt_disable_state, flags);
> +}
> +
> +static __always_inline void __local_interrupt_enable(void)
> +{
> + unsigned long flags = raw_cpu_read(local_interrupt_disable_state);
> +
> + local_irq_restore(flags);
> +}
> +
> +#ifndef INSTANTIATE_EXPORTED_INTERRUPT_DISABLE
> +static __always_inline void _local_interrupt_disable(void)
> +{
> + __local_interrupt_disable();
> +}
> +
> +static __always_inline void _local_interrupt_enable(void)
> +{
> + __local_interrupt_enable();
> +}
> +#else
> +extern void _local_interrupt_disable(void);
> +extern void _local_interrupt_enable(void);
> +#endif
> +
> +#else /* !MODULE */
> +extern void _local_interrupt_disable(void);
> +extern void _local_interrupt_enable(void);
> +#endif /* !MODULE */
> +
> +static inline void local_interrupt_disable(void)
> +{
> + int new_count;
> +
> + WARN_ON_ONCE(in_nmi());
> +
> + new_count = hardirq_disable_enter();
> +
> + /* Interrupts can happen here, but it's OK, see __irq_exit_rcu(). */
> +
> + if ((new_count & HARDIRQ_DISABLE_MASK) == HARDIRQ_DISABLE_OFFSET)
> + _local_interrupt_disable();
> +}
Maximum nesting possible is 256 right? Whats is stopping here to do more than that?
Should there be a warn_on?
> +
> +static inline void local_interrupt_enable(void)
> +{
> + int new_count;
> +
> + new_count = hardirq_disable_exit();
> +
> + if ((new_count & HARDIRQ_DISABLE_MASK) == 0)
> + _local_interrupt_enable();
> +}
> +
> +#endif /* !__LINUX_INTERRUPT_RC_H */
next prev parent reply other threads:[~2026-08-04 20:52 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-04 16:14 [PATCH v4 00/17] Refcounted interrupt disable and SpinLockIrq for Rust Boqun Feng
2026-08-04 16:14 ` [PATCH v4 01/17] preempt: Track NMI nesting to separate per-CPU counter Boqun Feng
2026-08-04 16:14 ` [PATCH v4 02/17] preempt: Introduce HARDIRQ_DISABLE_BITS Boqun Feng
2026-08-05 6:31 ` Peter Zijlstra
2026-08-05 6:59 ` Boqun Feng
2026-08-04 16:14 ` [PATCH v4 03/17] preempt: Introduce __preempt_count_{sub,add}_return() Boqun Feng
2026-08-04 16:14 ` [PATCH v4 04/17] openrisc: Include <linux/cpumask.h> in smp.h Boqun Feng
2026-08-04 16:14 ` [PATCH v4 05/17] irq & spin_lock: Add counted interrupt disabling/enabling Boqun Feng
2026-08-04 18:20 ` Boqun Feng
2026-08-04 18:26 ` [PATCH v4.1 " Boqun Feng
2026-08-04 20:51 ` Shrikanth Hegde [this message]
2026-08-04 21:08 ` [PATCH v4 " Boqun Feng
2026-08-05 6:36 ` Peter Zijlstra
2026-08-05 7:07 ` Boqun Feng
2026-08-05 7:09 ` Shrikanth Hegde
2026-08-05 7:19 ` Boqun Feng
2026-08-05 13:53 ` Boqun Feng
2026-08-05 14:10 ` Shrikanth Hegde
2026-08-05 14:20 ` Boqun Feng
2026-08-05 14:56 ` Shrikanth Hegde
2026-08-05 15:11 ` Boqun Feng
2026-08-05 16:53 ` Shrikanth Hegde
2026-08-05 17:38 ` Boqun Feng
2026-08-04 16:14 ` [PATCH v4 06/17] irq: Add KUnit test for refcounted interrupt enable/disable Boqun Feng
2026-08-04 16:14 ` [PATCH v4 07/17] locking: Switch to _irq_{disable,enable}() variants in cleanup guards Boqun Feng
2026-08-04 16:14 ` [PATCH v4 08/17] sched: Remove the unused preempt_offset parameter of __cant_sleep() Boqun Feng
2026-08-04 16:14 ` [PATCH v4 09/17] sched: Avoid signed comparison of preempt_count() in __cant_migrate() Boqun Feng
2026-08-04 16:14 ` [PATCH v4 10/17] preempt: Introduce HAS_SEPARATE_PREEMPT_RESCHED_BITS Boqun Feng
2026-08-04 20:11 ` Shrikanth Hegde
2026-08-05 6:54 ` Boqun Feng
2026-08-05 7:15 ` Shrikanth Hegde
2026-08-05 7:27 ` Boqun Feng
2026-08-06 0:58 ` Boqun Feng
2026-08-04 21:09 ` Shrikanth Hegde
2026-08-04 23:14 ` Boqun Feng
2026-08-04 16:14 ` [PATCH v4 11/17] arm64: sched/preempt: Enable HAS_SEPARATE_PREEMPT_RESCHED_BITS Boqun Feng
2026-08-04 16:14 ` [PATCH v4 12/17] s390/preempt: " Boqun Feng
2026-08-04 20:27 ` Shrikanth Hegde
2026-08-05 9:42 ` Peter Zijlstra
2026-08-05 12:37 ` Shrikanth Hegde
2026-08-04 16:14 ` [PATCH v4 13/17] rust: Introduce interrupt module Boqun Feng
2026-08-04 16:14 ` [PATCH v4 14/17] rust: helper: Add spin_{un,}lock_irq_{enable,disable}() helpers Boqun Feng
2026-08-04 16:14 ` [PATCH v4 15/17] rust: sync: Use super::* in spinlock.rs Boqun Feng
2026-08-04 16:14 ` [PATCH v4 16/17] rust: sync: Add SpinLockIrq Boqun Feng
2026-08-04 16:14 ` [PATCH v4 17/17] rust: sync: Introduce SpinLockIrq::lock_with() and friends Boqun Feng
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=2fc01d90-e081-4ddf-a842-b68eda15ca9e@linux.ibm.com \
--to=sshegde@linux.ibm.com \
--cc=aliceryhl@google.com \
--cc=boqun@kernel.org \
--cc=dakr@kernel.org \
--cc=daniel.almeida@collabora.com \
--cc=gary@garyguo.net \
--cc=linux-kernel@vger.kernel.org \
--cc=longman@redhat.com \
--cc=lyude@redhat.com \
--cc=mingo@kernel.org \
--cc=ojeda@kernel.org \
--cc=peterz@infradead.org \
--cc=rust-for-linux@vger.kernel.org \
--cc=will@kernel.org \
--cc=work@onurozkan.dev \
/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