linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Peter Oskolkov <posk@posk.io>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Peter Oskolkov <posk@google.com>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	"Paul E . McKenney" <paulmck@kernel.org>,
	Boqun Feng <boqun.feng@gmail.com>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Paul Turner <pjt@google.com>,
	Chris Kennelly <ckennelly@google.com>
Subject: Re: [PATCH 1/2 v2] rseq/membarrier: add MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU
Date: Fri, 7 Aug 2020 10:50:42 -0700	[thread overview]
Message-ID: <CAFTs51VHC5nyAb7bf-jvuYDRwLS5kWP98T336Uu7h+RMFb0ZqA@mail.gmail.com> (raw)
In-Reply-To: <20200807133755.GJ2674@hirez.programming.kicks-ass.net>

On Fri, Aug 7, 2020 at 6:38 AM <peterz@infradead.org> wrote:
>

[...]

> I'm thinking even this is a problem, we can end up sending IPIs to CPUs
> outside out partition (they might be NOHZ_FULL) and that's a no-no too.
>
> Something like so perhaps... that really limits it to CPUs that match
> our mm.

Thanks for the suggestion - I'll prepare a v3 based on your and
Mathieu's feedback.

>
> diff --git a/include/linux/sched/mm.h b/include/linux/sched/mm.h
> index 6be66f52a2ad..bee5e98e6774 100644
> --- a/include/linux/sched/mm.h
> +++ b/include/linux/sched/mm.h
> @@ -356,6 +356,7 @@ enum {
>
>  enum {
>         MEMBARRIER_FLAG_SYNC_CORE       = (1U << 0),
> +       MEMBARRIER_FLAG_RSEQ            = (1U << 1),
>  };
>
>  #ifdef CONFIG_ARCH_HAS_MEMBARRIER_CALLBACKS
> diff --git a/kernel/sched/membarrier.c b/kernel/sched/membarrier.c
> index 168479a7d61b..4d9b22c2f5e2 100644
> --- a/kernel/sched/membarrier.c
> +++ b/kernel/sched/membarrier.c
> @@ -27,6 +27,11 @@
>
>  static void ipi_mb(void *info)
>  {
> +       int *flags = info;
> +
> +       if (flags && (*flags & MEMBARRIER_FLAG_RSEQ))
> +               rseq_preempt(current);
> +
>         smp_mb();       /* IPIs should be serializing but paranoid. */
>  }
>
> @@ -129,11 +134,11 @@ static int membarrier_global_expedited(void)
>         return 0;
>  }
>
> -static int membarrier_private_expedited(int flags)
> +static int membarrier_private_expedited(int flags, int cpu_id)
>  {
> -       int cpu;
> -       cpumask_var_t tmpmask;
>         struct mm_struct *mm = current->mm;
> +       cpumask_var_t tmpmask;
> +       int cpu;
>
>         if (flags & MEMBARRIER_FLAG_SYNC_CORE) {
>                 if (!IS_ENABLED(CONFIG_ARCH_HAS_MEMBARRIER_SYNC_CORE))
> @@ -174,6 +179,10 @@ static int membarrier_private_expedited(int flags)
>                  */
>                 if (cpu == raw_smp_processor_id())
>                         continue;
> +
> +               if (cpu_id >= 0 && cpu != cpu_id)
> +                       continue;
> +
>                 p = rcu_dereference(cpu_rq(cpu)->curr);
>                 if (p && p->mm == mm)
>                         __cpumask_set_cpu(cpu, tmpmask);
> @@ -181,7 +190,7 @@ static int membarrier_private_expedited(int flags)
>         rcu_read_unlock();
>
>         preempt_disable();
> -       smp_call_function_many(tmpmask, ipi_mb, NULL, 1);
> +       smp_call_function_many(tmpmask, ipi_mb, &flags, 1);
>         preempt_enable();
>
>         free_cpumask_var(tmpmask);
> @@ -362,11 +371,13 @@ SYSCALL_DEFINE2(membarrier, int, cmd, int, flags)
>         case MEMBARRIER_CMD_REGISTER_GLOBAL_EXPEDITED:
>                 return membarrier_register_global_expedited();
>         case MEMBARRIER_CMD_PRIVATE_EXPEDITED:
> -               return membarrier_private_expedited(0);
> +               return membarrier_private_expedited(0, -1);
>         case MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED:
>                 return membarrier_register_private_expedited(0);
>         case MEMBARRIER_CMD_PRIVATE_EXPEDITED_SYNC_CORE:
> -               return membarrier_private_expedited(MEMBARRIER_FLAG_SYNC_CORE);
> +               return membarrier_private_expedited(MEMBARRIER_FLAG_SYNC_CORE, -1);
> +       case MEMBERRIER_CMD_PRIVATE_EXPEDITED_RSEQ:
> +               return membarrier_private_expedited(MEMBARRIER_FLAG_RSEQ, flags);
>         case MEMBARRIER_CMD_REGISTER_PRIVATE_EXPEDITED_SYNC_CORE:
>                 return membarrier_register_private_expedited(MEMBARRIER_FLAG_SYNC_CORE);
>         default:

      reply	other threads:[~2020-08-07 17:50 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-06 17:05 [PATCH 1/2 v2] rseq/membarrier: add MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU Peter Oskolkov
2020-08-06 17:05 ` [PATCH 2/2 v2] rseq/selftests: test MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU Peter Oskolkov
2020-08-07  0:27   ` Boqun Feng
2020-08-07 12:54     ` Mathieu Desnoyers
2020-08-07 17:55     ` Peter Oskolkov
2020-08-07 18:25       ` Mathieu Desnoyers
2020-08-07 18:47         ` Peter Oskolkov
2020-08-07 20:29           ` Mathieu Desnoyers
2020-08-07 13:37 ` [PATCH 1/2 v2] rseq/membarrier: add MEMBARRIER_CMD_PRIVATE_RESTART_RSEQ_ON_CPU peterz
2020-08-07 17:50   ` Peter Oskolkov [this message]

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=CAFTs51VHC5nyAb7bf-jvuYDRwLS5kWP98T336Uu7h+RMFb0ZqA@mail.gmail.com \
    --to=posk@posk.io \
    --cc=boqun.feng@gmail.com \
    --cc=ckennelly@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=paulmck@kernel.org \
    --cc=peterz@infradead.org \
    --cc=pjt@google.com \
    --cc=posk@google.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).