The Linux Kernel Mailing List
 help / color / mirror / Atom feed
From: Peter Zijlstra <peterz@infradead.org>
To: Lyude Paul <lyude@redhat.com>
Cc: rust-for-linux@vger.kernel.org, linux-kernel@vger.kernel.org,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Boqun Feng" <boqun.feng@gmail.com>,
	"Daniel Almeida" <daniel.almeida@collabora.com>,
	"Miguel Ojeda" <ojeda@kernel.org>,
	"Alex Gaynor" <alex.gaynor@gmail.com>,
	"Gary Guo" <gary@garyguo.net>,
	"Björn Roy Baron" <bjorn3_gh@protonmail.com>,
	"Benno Lossin" <lossin@kernel.org>,
	"Andreas Hindborg" <a.hindborg@kernel.org>,
	"Alice Ryhl" <aliceryhl@google.com>,
	"Trevor Gross" <tmgross@umich.edu>,
	"Danilo Krummrich" <dakr@kernel.org>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	"Ingo Molnar" <mingo@redhat.com>, "Will Deacon" <will@kernel.org>,
	"Waiman Long" <longman@redhat.com>,
	"Joel Fernandes" <joelagnelf@nvidia.com>
Subject: Re: [PATCH v17 02/16] preempt: Track NMI nesting to separate per-CPU counter
Date: Tue, 3 Feb 2026 13:15:20 +0100	[thread overview]
Message-ID: <20260203121520.GN1282955@noisy.programming.kicks-ass.net> (raw)
In-Reply-To: <20260121223933.1568682-3-lyude@redhat.com>

On Wed, Jan 21, 2026 at 05:39:05PM -0500, Lyude Paul wrote:

>  #define __nmi_enter()						\
>  	do {							\
>  		lockdep_off();					\
>  		arch_nmi_enter();				\
> -		BUG_ON(in_nmi() == NMI_MASK);			\
> -		__preempt_count_add(NMI_OFFSET + HARDIRQ_OFFSET);	\
> +		BUG_ON(__this_cpu_read(nmi_nesting) == UINT_MAX);	\
> +		__this_cpu_inc(nmi_nesting);			\
> +		__preempt_count_add(HARDIRQ_OFFSET);		\
> +		preempt_count_set(preempt_count() | NMI_MASK);	\
>  	} while (0)
>  
>  #define nmi_enter()						\
> @@ -124,8 +128,12 @@ void irq_exit_rcu(void);
>  
>  #define __nmi_exit()						\
>  	do {							\
> +		unsigned int nesting;				\
>  		BUG_ON(!in_nmi());				\
> -		__preempt_count_sub(NMI_OFFSET + HARDIRQ_OFFSET);	\
> +		__preempt_count_sub(HARDIRQ_OFFSET);		\
> +		nesting = __this_cpu_dec_return(nmi_nesting);	\
> +		if (!nesting)					\
> +			__preempt_count_sub(NMI_OFFSET);	\
>  		arch_nmi_exit();				\
>  		lockdep_on();					\
>  	} while (0)

While not wrong like last time; it is pretty awful. 

preempt_count_set() is a cmpxchg() loop.

Would not something like so be better?

#define __nmi_enter()                                           \
        do {                                                    \
+               unsigned int _o = NMI_MASK + HARDIRQ_OFFSET;    \
                lockdep_off();                                  \
                arch_nmi_enter();                               \
-               BUG_ON(in_nmi() == NMI_MASK);                   \
-               __preempt_count_add(NMI_OFFSET + HARDIRQ_OFFSET);       \
+               BUG_ON(__this_cpu_read(nmi_nesting) == ~0U);    \
+               __this_cpu_inc(nmi_nesting);                    \
+               _o -= (preempt_count() & NMI_MASK);             \
+               __preempt_count_add(_o);                        \
        } while (0)


 #define __nmi_exit()                                           \
        do {                                                    \
+               unsigned int _o = HARDIRQ_OFFSET;               \
                BUG_ON(!in_nmi());                              \
-               __preempt_count_sub(NMI_OFFSET + HARDIRQ_OFFSET);       \
+               if (!__this_cpu_dec_return(nmi_nesting))        \
+                       _o += NMI_MASK;                         \
+               __preempt_count_sub(_o);                        \
                arch_nmi_exit();                                \
                lockdep_on();                                   \
        } while (0)


But I'm really somewhat sad that 64bit can't do better than this.



  parent reply	other threads:[~2026-02-03 12:15 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-21 22:39 [PATCH v17 00/16] Refcounted interrupts, SpinLockIrq for rust Lyude Paul
2026-01-21 22:39 ` [PATCH v17 01/16] preempt: Introduce HARDIRQ_DISABLE_BITS Lyude Paul
2026-01-21 22:39 ` [PATCH v17 02/16] preempt: Track NMI nesting to separate per-CPU counter Lyude Paul
2026-02-03 11:44   ` Peter Zijlstra
2026-02-06  1:22     ` Joel Fernandes
2026-02-03 12:15   ` Peter Zijlstra [this message]
2026-02-04 11:12     ` Peter Zijlstra
2026-02-04 12:32       ` Gary Guo
2026-02-04 13:00         ` Peter Zijlstra
2026-02-05 21:40       ` Boqun Feng
2026-02-05 22:17         ` Joel Fernandes
2026-02-06  0:50           ` Joel Fernandes
2026-02-06  1:14             ` Boqun Feng
2026-02-06  1:24               ` Joel Fernandes
2026-02-06  2:51                 ` Boqun Feng
2026-02-06  8:13                   ` Joel Fernandes
2026-02-06 15:28                     ` Boqun Feng
2026-02-06 16:00                       ` Joel Fernandes
2026-02-06 16:16                         ` Boqun Feng
2026-02-07 22:11                           ` Joel Fernandes
2026-02-06  8:42                 ` Peter Zijlstra
2026-02-05 22:07       ` Boqun Feng
2026-02-06  8:45         ` Peter Zijlstra
2026-01-21 22:39 ` [PATCH v17 03/16] preempt: Introduce __preempt_count_{sub, add}_return() Lyude Paul
2026-01-21 22:39 ` [PATCH v17 04/16] openrisc: Include <linux/cpumask.h> in smp.h Lyude Paul
2026-01-21 22:39 ` [PATCH v17 05/16] irq & spin_lock: Add counted interrupt disabling/enabling Lyude Paul
2026-01-21 22:39 ` [PATCH v17 06/16] irq: Add KUnit test for refcounted interrupt enable/disable Lyude Paul
2026-01-30  7:43   ` David Gow
2026-01-21 22:39 ` [PATCH v17 07/16] rust: Introduce interrupt module Lyude Paul
2026-01-21 22:39 ` [PATCH v17 08/16] rust: helper: Add spin_{un,}lock_irq_{enable,disable}() helpers Lyude Paul
2026-01-26 13:25   ` Gary Guo
2026-01-21 22:39 ` [PATCH v17 09/16] rust: sync: Add SpinLockIrq Lyude Paul
2026-01-23 22:26   ` Benno Lossin
2026-01-21 22:39 ` [PATCH v17 10/16] rust: sync: Introduce lock::Lock::lock_with() and friends Lyude Paul
2026-01-22 11:56   ` kernel test robot
2026-01-23 22:55   ` Benno Lossin
2026-01-26 13:31   ` Gary Guo
2026-01-21 22:39 ` [PATCH v17 11/16] rust: sync: Expose lock::Backend Lyude Paul
2026-01-23 22:56   ` Benno Lossin
2026-01-21 22:39 ` [PATCH v17 12/16] rust: sync: lock/global: Rename B to G in trait bounds Lyude Paul
2026-01-21 22:39 ` [PATCH v17 13/16] rust: sync: Add a lifetime parameter to lock::global::GlobalGuard Lyude Paul
2026-01-21 22:39 ` [PATCH v17 14/16] rust: sync: lock/global: Add Backend parameter to GlobalGuard Lyude Paul
2026-01-21 22:39 ` [PATCH v17 15/16] rust: sync: lock/global: Add ContextualBackend support to GlobalLock Lyude Paul
2026-01-21 22:39 ` [PATCH v17 16/16] locking: Switch to _irq_{disable,enable}() variants in cleanup guards Lyude Paul
2026-01-26 13:24 ` [PATCH v17 00/16] Refcounted interrupts, SpinLockIrq for rust Gary Guo
2026-01-26 16:17 ` Boqun Feng
2026-02-03  0:36   ` 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=20260203121520.GN1282955@noisy.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=a.hindborg@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=alex.gaynor@gmail.com \
    --cc=aliceryhl@google.com \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun.feng@gmail.com \
    --cc=dakr@kernel.org \
    --cc=daniel.almeida@collabora.com \
    --cc=gary@garyguo.net \
    --cc=joelagnelf@nvidia.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=longman@redhat.com \
    --cc=lossin@kernel.org \
    --cc=lyude@redhat.com \
    --cc=mingo@redhat.com \
    --cc=ojeda@kernel.org \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=tmgross@umich.edu \
    --cc=will@kernel.org \
    /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