mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
To: Thomas Gleixner <tglx@linutronix.de>,
	LKML <linux-kernel@vger.kernel.org>
Cc: Michael Jeanson <mjeanson@efficios.com>,
	Jens Axboe <axboe@kernel.dk>, Paolo Bonzini <pbonzini@redhat.com>,
	Sean Christopherson <seanjc@google.com>,
	Wei Liu <wei.liu@kernel.org>, Dexuan Cui <decui@microsoft.com>,
	Peter Zijlstra <peterz@infradead.org>,
	"Paul E. McKenney" <paulmck@kernel.org>,
	Boqun Feng <boqun.feng@gmail.com>,
	x86@kernel.org, Arnd Bergmann <arnd@arndb.de>,
	Heiko Carstens <hca@linux.ibm.com>,
	Christian Borntraeger <borntraeger@linux.ibm.com>,
	Sven Schnelle <svens@linux.ibm.com>,
	Huacai Chen <chenhuacai@kernel.org>,
	Paul Walmsley <paul.walmsley@sifive.com>,
	Palmer Dabbelt <palmer@dabbelt.com>
Subject: Re: [patch V4 07/36] rseq, virt: Retrigger RSEQ after vcpu_run()
Date: Tue, 9 Sep 2025 09:21:13 -0400	[thread overview]
Message-ID: <ee2c27ed-1c72-4fff-8798-1ad20016da95@efficios.com> (raw)
In-Reply-To: <20250908212925.708769718@linutronix.de>

On 2025-09-08 17:31, Thomas Gleixner wrote:
> Hypervisors invoke resume_user_mode_work() before entering the guest, which
> clears TIF_NOTIFY_RESUME. The @regs argument is NULL as there is no user
> space context available to them, so the rseq notify handler skips
> inspecting the critical section, but updates the CPU/MM CID values
> unconditionally so that the eventual pending rseq event is not lost on the
> way to user space.
> 
> This is a pointless exercise as the task might be rescheduled before
> actually returning to user space and it creates unnecessary work in the
> vcpu_run() loops.
> 
> It's way more efficient to ignore that invocation based on @regs == NULL
> and let the hypervisors re-raise TIF_NOTIFY_RESUME after returning from the
> vcpu_run() loop before returning from the ioctl().
> 
> This ensures that a pending RSEQ update is not lost and the IDs are updated
> before returning to user space.
> 
> Once the RSEQ handling is decoupled from TIF_NOTIFY_RESUME, this turns into
> a NOOP.

Reviewed-by: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>

> 
> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Sean Christopherson <seanjc@google.com>
> Cc: Wei Liu <wei.liu@kernel.org>
> Cc: Dexuan Cui <decui@microsoft.com>
> ---
> V3: Add the missing rseq.h include for HV - 0-day
> ---
>   drivers/hv/mshv_root_main.c |    3 +
>   include/linux/rseq.h        |   17 +++++++++
>   kernel/rseq.c               |   76 +++++++++++++++++++++++---------------------
>   virt/kvm/kvm_main.c         |    3 +
>   4 files changed, 63 insertions(+), 36 deletions(-)
> 
> --- a/drivers/hv/mshv_root_main.c
> +++ b/drivers/hv/mshv_root_main.c
> @@ -28,6 +28,7 @@
>   #include <linux/crash_dump.h>
>   #include <linux/panic_notifier.h>
>   #include <linux/vmalloc.h>
> +#include <linux/rseq.h>
>   
>   #include "mshv_eventfd.h"
>   #include "mshv.h"
> @@ -585,6 +586,8 @@ static long mshv_run_vp_with_root_schedu
>   		}
>   	} while (!vp->run.flags.intercept_suspend);
>   
> +	rseq_virt_userspace_exit();
> +
>   	return ret;
>   }
>   
> --- a/include/linux/rseq.h
> +++ b/include/linux/rseq.h
> @@ -38,6 +38,22 @@ static __always_inline void rseq_exit_to
>   }
>   
>   /*
> + * KVM/HYPERV invoke resume_user_mode_work() before entering guest mode,
> + * which clears TIF_NOTIFY_RESUME. To avoid updating user space RSEQ in
> + * that case just to do it eventually again before returning to user space,
> + * the entry resume_user_mode_work() invocation is ignored as the register
> + * argument is NULL.
> + *
> + * After returning from guest mode, they have to invoke this function to
> + * re-raise TIF_NOTIFY_RESUME if necessary.
> + */
> +static inline void rseq_virt_userspace_exit(void)
> +{
> +	if (current->rseq_event_pending)
> +		set_tsk_thread_flag(current, TIF_NOTIFY_RESUME);
> +}
> +
> +/*
>    * If parent process has a registered restartable sequences area, the
>    * child inherits. Unregister rseq for a clone with CLONE_VM set.
>    */
> @@ -68,6 +84,7 @@ static inline void rseq_execve(struct ta
>   static inline void rseq_handle_notify_resume(struct ksignal *ksig, struct pt_regs *regs) { }
>   static inline void rseq_signal_deliver(struct ksignal *ksig, struct pt_regs *regs) { }
>   static inline void rseq_sched_switch_event(struct task_struct *t) { }
> +static inline void rseq_virt_userspace_exit(void) { }
>   static inline void rseq_fork(struct task_struct *t, unsigned long clone_flags) { }
>   static inline void rseq_execve(struct task_struct *t) { }
>   static inline void rseq_exit_to_user_mode(void) { }
> --- a/kernel/rseq.c
> +++ b/kernel/rseq.c
> @@ -422,50 +422,54 @@ void __rseq_handle_notify_resume(struct
>   {
>   	struct task_struct *t = current;
>   	int ret, sig;
> +	bool event;
> +
> +	/*
> +	 * If invoked from hypervisors before entering the guest via
> +	 * resume_user_mode_work(), then @regs is a NULL pointer.
> +	 *
> +	 * resume_user_mode_work() clears TIF_NOTIFY_RESUME and re-raises
> +	 * it before returning from the ioctl() to user space when
> +	 * rseq_event.sched_switch is set.
> +	 *
> +	 * So it's safe to ignore here instead of pointlessly updating it
> +	 * in the vcpu_run() loop.
> +	 */
> +	if (!regs)
> +		return;
>   
>   	if (unlikely(t->flags & PF_EXITING))
>   		return;
>   
>   	/*
> -	 * If invoked from hypervisors or IO-URING, then @regs is a NULL
> -	 * pointer, so fixup cannot be done. If the syscall which led to
> -	 * this invocation was invoked inside a critical section, then it
> -	 * will either end up in this code again or a possible violation of
> -	 * a syscall inside a critical region can only be detected by the
> -	 * debug code in rseq_syscall() in a debug enabled kernel.
> +	 * Read and clear the event pending bit first. If the task
> +	 * was not preempted or migrated or a signal is on the way,
> +	 * there is no point in doing any of the heavy lifting here
> +	 * on production kernels. In that case TIF_NOTIFY_RESUME
> +	 * was raised by some other functionality.
> +	 *
> +	 * This is correct because the read/clear operation is
> +	 * guarded against scheduler preemption, which makes it CPU
> +	 * local atomic. If the task is preempted right after
> +	 * re-enabling preemption then TIF_NOTIFY_RESUME is set
> +	 * again and this function is invoked another time _before_
> +	 * the task is able to return to user mode.
> +	 *
> +	 * On a debug kernel, invoke the fixup code unconditionally
> +	 * with the result handed in to allow the detection of
> +	 * inconsistencies.
>   	 */
> -	if (regs) {
> -		/*
> -		 * Read and clear the event pending bit first. If the task
> -		 * was not preempted or migrated or a signal is on the way,
> -		 * there is no point in doing any of the heavy lifting here
> -		 * on production kernels. In that case TIF_NOTIFY_RESUME
> -		 * was raised by some other functionality.
> -		 *
> -		 * This is correct because the read/clear operation is
> -		 * guarded against scheduler preemption, which makes it CPU
> -		 * local atomic. If the task is preempted right after
> -		 * re-enabling preemption then TIF_NOTIFY_RESUME is set
> -		 * again and this function is invoked another time _before_
> -		 * the task is able to return to user mode.
> -		 *
> -		 * On a debug kernel, invoke the fixup code unconditionally
> -		 * with the result handed in to allow the detection of
> -		 * inconsistencies.
> -		 */
> -		bool event;
> -
> -		scoped_guard(RSEQ_EVENT_GUARD) {
> -			event = t->rseq_event_pending;
> -			t->rseq_event_pending = false;
> -		}
> +	scoped_guard(RSEQ_EVENT_GUARD) {
> +		event = t->rseq_event_pending;
> +		t->rseq_event_pending = false;
> +	}
>   
> -		if (IS_ENABLED(CONFIG_DEBUG_RSEQ) || event) {
> -			ret = rseq_ip_fixup(regs, event);
> -			if (unlikely(ret < 0))
> -				goto error;
> -		}
> +	if (IS_ENABLED(CONFIG_DEBUG_RSEQ) || event) {
> +		ret = rseq_ip_fixup(regs, event);
> +		if (unlikely(ret < 0))
> +			goto error;
>   	}
> +
>   	if (unlikely(rseq_update_cpu_node_id(t)))
>   		goto error;
>   	return;
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -49,6 +49,7 @@
>   #include <linux/lockdep.h>
>   #include <linux/kthread.h>
>   #include <linux/suspend.h>
> +#include <linux/rseq.h>
>   
>   #include <asm/processor.h>
>   #include <asm/ioctl.h>
> @@ -4466,6 +4467,8 @@ static long kvm_vcpu_ioctl(struct file *
>   		r = kvm_arch_vcpu_ioctl_run(vcpu);
>   		vcpu->wants_to_run = false;
>   
> +		rseq_virt_userspace_exit();
> +
>   		trace_kvm_userspace_exit(vcpu->run->exit_reason, r);
>   		break;
>   	}
> 


-- 
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com

  parent reply	other threads:[~2025-09-09 13:21 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-08 21:31 [patch V4 00/36] rseq: Optimize exit to user space Thomas Gleixner
2025-09-08 21:31 ` [patch V4 01/36] rseq: Avoid pointless evaluation in __rseq_notify_resume() Thomas Gleixner
2025-09-08 21:31 ` [patch V4 02/36] rseq: Condense the inline stubs Thomas Gleixner
2025-09-08 21:31 ` [patch V4 03/36] rseq: Move algorithm comment to top Thomas Gleixner
2025-09-08 21:31 ` [patch V4 04/36] rseq: Remove the ksig argument from rseq_handle_notify_resume() Thomas Gleixner
2025-09-08 21:31 ` [patch V4 05/36] rseq: Simplify registration Thomas Gleixner
2025-09-08 21:31 ` [patch V4 06/36] rseq: Simplify the event notification Thomas Gleixner
2025-09-09 13:18   ` Mathieu Desnoyers
2025-09-08 21:31 ` [patch V4 07/36] rseq, virt: Retrigger RSEQ after vcpu_run() Thomas Gleixner
2025-09-09  0:00   ` Sean Christopherson
2025-09-09 12:10     ` Thomas Gleixner
2025-09-09 13:21   ` Mathieu Desnoyers [this message]
2025-09-08 21:31 ` [patch V4 08/36] rseq: Avoid CPU/MM CID updates when no event pending Thomas Gleixner
2025-09-09 13:25   ` Mathieu Desnoyers
2025-09-08 21:31 ` [patch V4 09/36] rseq: Introduce struct rseq_data Thomas Gleixner
2025-09-09 13:30   ` Mathieu Desnoyers
2025-09-12 20:44     ` Thomas Gleixner
2025-09-12 21:33       ` Mathieu Desnoyers
2025-09-08 21:31 ` [patch V4 10/36] entry: Cleanup header Thomas Gleixner
2025-09-08 21:31 ` [patch V4 11/36] entry: Remove syscall_enter_from_user_mode_prepare() Thomas Gleixner
2025-09-09 13:33   ` Mathieu Desnoyers
2025-09-08 21:31 ` [patch V4 12/36] entry: Inline irqentry_enter/exit_from/to_user_mode() Thomas Gleixner
2025-09-09 13:38   ` Mathieu Desnoyers
2025-09-09 14:10     ` Thomas Gleixner
2025-09-09 14:59       ` Mathieu Desnoyers
2025-09-08 21:31 ` [patch V4 13/36] sched: Move MM CID related functions to sched.h Thomas Gleixner
2025-09-08 21:31 ` [patch V4 14/36] rseq: Cache CPU ID and MM CID values Thomas Gleixner
2025-09-09 13:43   ` Mathieu Desnoyers
2025-09-09 14:13     ` Thomas Gleixner
2025-09-09 15:01       ` Mathieu Desnoyers
2025-09-08 21:31 ` [patch V4 15/36] rseq: Record interrupt from user space Thomas Gleixner
2025-09-09 13:53   ` Mathieu Desnoyers
2025-09-09 14:17     ` Thomas Gleixner
2025-09-09 15:05       ` Mathieu Desnoyers
2025-09-08 21:31 ` [patch V4 16/36] rseq: Provide tracepoint wrappers for inline code Thomas Gleixner
2025-09-08 21:31 ` [patch V4 17/36] rseq: Expose lightweight statistics in debugfs Thomas Gleixner
2025-09-08 21:32 ` [patch V4 18/36] rseq: Provide static branch for runtime debugging Thomas Gleixner
2025-09-08 21:32 ` [patch V4 19/36] rseq: Provide and use rseq_update_user_cs() Thomas Gleixner
2025-09-09 15:11   ` Mathieu Desnoyers
2025-09-08 21:32 ` [patch V4 20/36] rseq: Replace the original debug implementation Thomas Gleixner
2025-09-08 21:32 ` [patch V4 21/36] rseq: Make exit debugging static branch based Thomas Gleixner
2025-09-08 21:32 ` [patch V4 22/36] rseq: Use static branch for syscall exit debug when GENERIC_IRQ_ENTRY=y Thomas Gleixner
2025-09-08 21:32 ` [patch V4 23/36] rseq: Provide and use rseq_set_ids() Thomas Gleixner
2025-09-11 13:40   ` Mathieu Desnoyers
2025-09-11 16:02     ` Thomas Gleixner
2025-09-11 17:13       ` Mathieu Desnoyers
2025-09-08 21:32 ` [patch V4 24/36] rseq: Separate the signal delivery path Thomas Gleixner
2025-09-08 21:32 ` [patch V4 25/36] rseq: Rework the TIF_NOTIFY handler Thomas Gleixner
2025-09-08 21:32 ` [patch V4 26/36] rseq: Optimize event setting Thomas Gleixner
2025-09-11 14:03   ` Mathieu Desnoyers
2025-09-11 16:06     ` Thomas Gleixner
2025-09-11 17:15       ` Mathieu Desnoyers
2025-09-12  6:58         ` Thomas Gleixner
2025-09-08 21:32 ` [patch V4 27/36] rseq: Implement fast path for exit to user Thomas Gleixner
2025-09-11 14:27   ` Mathieu Desnoyers
2025-09-11 16:08     ` Thomas Gleixner
2025-09-08 21:32 ` [patch V4 28/36] rseq: Switch to fast path processing on " Thomas Gleixner
2025-09-11 14:44   ` Mathieu Desnoyers
2025-09-11 14:45     ` Mathieu Desnoyers
2025-09-11 16:50       ` Thomas Gleixner
2025-09-11 16:47     ` Thomas Gleixner
2025-09-11 20:00       ` Mathieu Desnoyers
2025-09-12 14:22         ` Thomas Gleixner
2025-09-12 15:44           ` Mathieu Desnoyers
2025-09-08 21:32 ` [patch V4 29/36] entry: Split up exit_to_user_mode_prepare() Thomas Gleixner
2025-09-08 21:32 ` [patch V4 30/36] rseq: Split up rseq_exit_to_user_mode() Thomas Gleixner
2025-09-08 21:32 ` [patch V4 31/36] asm-generic: Provide generic TIF infrastructure Thomas Gleixner
2025-09-17  6:16   ` [tip: core/core] " tip-bot2 for Thomas Gleixner
2025-09-08 21:32 ` [patch V4 32/36] x86: Use generic TIF bits Thomas Gleixner
2025-09-17  6:16   ` [tip: core/core] " tip-bot2 for Thomas Gleixner
2025-09-08 21:32 ` [patch V4 33/36] s390: " Thomas Gleixner
2025-09-11  9:11   ` Sven Schnelle
2025-09-11 11:03   ` Heiko Carstens
2025-09-17  6:16   ` [tip: core/core] " tip-bot2 for Thomas Gleixner
2025-09-08 21:32 ` [patch V4 34/36] loongarch: " Thomas Gleixner
2025-09-17  6:16   ` [tip: core/core] " tip-bot2 for Thomas Gleixner
2025-09-08 21:32 ` [patch V4 35/36] riscv: " Thomas Gleixner
2025-09-17  6:16   ` [tip: core/core] " tip-bot2 for Thomas Gleixner
2025-09-08 21:32 ` [patch V4 36/36] rseq: Switch to TIF_RSEQ if supported Thomas Gleixner
2025-09-10 13:55 ` [patch V4 00/36] rseq: Optimize exit to user space Jens Axboe
2025-09-10 14:45   ` Michael Jeanson
2025-09-10 15:34     ` Jens Axboe
2025-09-10 14:54   ` Thomas Gleixner

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=ee2c27ed-1c72-4fff-8798-1ad20016da95@efficios.com \
    --to=mathieu.desnoyers@efficios.com \
    --cc=arnd@arndb.de \
    --cc=axboe@kernel.dk \
    --cc=boqun.feng@gmail.com \
    --cc=borntraeger@linux.ibm.com \
    --cc=chenhuacai@kernel.org \
    --cc=decui@microsoft.com \
    --cc=hca@linux.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mjeanson@efficios.com \
    --cc=palmer@dabbelt.com \
    --cc=paul.walmsley@sifive.com \
    --cc=paulmck@kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=peterz@infradead.org \
    --cc=seanjc@google.com \
    --cc=svens@linux.ibm.com \
    --cc=tglx@linutronix.de \
    --cc=wei.liu@kernel.org \
    --cc=x86@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

all inboxes | Powered by JetHome®