Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.120.0.1.13\))
Subject: Re: [PATCH v2] perf/core: fix userpage->time_enabled of inactive
 events
From:   Athira Rajeev <atrajeev@linux.vnet.ibm.com>
In-Reply-To: <20210924012800.2461781-1-songliubraving@fb.com>
Date:   Wed, 29 Sep 2021 22:34:41 +0530
Cc:     LKML <linux-kernel@vger.kernel.org>,
        Arnaldo Carvalho de Melo <acme@kernel.org>,
        Peter Zijlstra <peterz@infradead.org>,
        Ingo Molnar <mingo@redhat.com>, kernel-team@fb.com,
        Stephane Eranian <eranian@google.com>,
        Lucian Grijincu <lucian@fb.com>,
        Madhavan Srinivasan <maddy@linux.vnet.ibm.com>,
        kajoljain <kjain@linux.ibm.com>
Content-Transfer-Encoding: quoted-printable
Message-Id: <EEA8D03E-14A7-4A8B-8C99-B68CACEC25D2@linux.vnet.ibm.com>
References: <20210924012800.2461781-1-songliubraving@fb.com>
To:     Song Liu <songliubraving@fb.com>
Precedence: bulk


> On 24-Sep-2021, at 6:58 AM, Song Liu <songliubraving@fb.com> wrote:
>=20
> Users of rdpmc rely on the mmapped user page to calculate accurate
> time_enabled. Currently, userpage->time_enabled is only updated when =
the
> event is added to the pmu. As a result, inactive event (due to counter
> multiplexing) does not have accurate userpage->time_enabled. This can
> be reproduced with something like:
>=20
>   /* open 20 task perf_event "cycles", to create multiplexing */
>=20
>   fd =3D perf_event_open();  /* open task perf_event "cycles" */
>   userpage =3D mmap(fd);     /* use mmap and rdmpc */
>=20
>   while (true) {
>     time_enabled_mmap =3D xxx; /* use logic in perf_event_mmap_page */
>     time_enabled_read =3D read(fd).time_enabled;
>     if (time_enabled_mmap > time_enabled_read)
>         BUG();
>   }
>=20
> Fix this by updating userpage for inactive events in merge_sched_in.
>=20
> Suggested-by: Peter Zijlstra (Intel) <peterz@infradead.org>
> Reported-and-tested-by: Lucian Grijincu <lucian@fb.com>
> Signed-off-by: Song Liu <songliubraving@fb.com>
> ---
> include/linux/perf_event.h |  4 +++-
> kernel/events/core.c       | 49 ++++++++++++++++++++++++++++++++++----
> 2 files changed, 48 insertions(+), 5 deletions(-)
>=20
> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index 2d510ad750edc..4aa52f7a48c16 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -683,7 +683,9 @@ struct perf_event {
> 	/*
> 	 * timestamp shadows the actual context timing but it can
> 	 * be safely used in NMI interrupt context. It reflects the
> -	 * context time as it was when the event was last scheduled in.
> +	 * context time as it was when the event was last scheduled in,
> +	 * or when ctx_sched_in failed to schedule the event because we
> +	 * run out of PMC.
> 	 *
> 	 * ctx_time already accounts for ctx->timestamp. Therefore to
> 	 * compute ctx_time for a sample, simply add perf_clock().
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 1cb1f9b8392e2..d73f986eef7b3 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -3707,6 +3707,46 @@ static noinline int visit_groups_merge(struct =
perf_cpu_context *cpuctx,
> 	return 0;
> }
>=20
> +static inline bool event_update_userpage(struct perf_event *event)
> +{
> +	/*
> +	 * Checking mmap_count to avoid unnecessary work. This does =
leave a
> +	 * corner case: if the event is enabled before mmap(), the first
> +	 * time the event gets scheduled is via:
> +	 *
> +	 *  __perf_event_enable (or __perf_install_in_context)
> +	 *      -> ctx_resched
> +	 *         -> perf_event_sched_in
> +	 *            -> ctx_sched_in
> +	 *
> +	 * with mmap_count of 0, so we will skip here. As a result,
> +	 * userpage->offset is not accurate after mmap and before the
> +	 * first rotation.
> +	 *
> +	 * To avoid the discrepancy of this window, the user space =
should
> +	 * mmap the event before enabling it.
> +	 */
> +	if (likely(!atomic_read(&event->mmap_count)))
> +		return false;
> +
> +	perf_event_update_time(event);
> +	perf_set_shadow_time(event, event->ctx);
> +	perf_event_update_userpage(event);
> +
> +	return true;
> +}
> +
> +static inline void group_update_userpage(struct perf_event =
*group_event)
> +{
> +	struct perf_event *event;
> +
> +	if (!event_update_userpage(group_event))
> +		return;
> +
> +	for_each_sibling_event(event, group_event)
> +		event_update_userpage(event);
> +}
> +
> static int merge_sched_in(struct perf_event *event, void *data)
> {
> 	struct perf_event_context *ctx =3D event->ctx;
> @@ -3725,14 +3765,15 @@ static int merge_sched_in(struct perf_event =
*event, void *data)
> 	}
>=20
> 	if (event->state =3D=3D PERF_EVENT_STATE_INACTIVE) {
> +		*can_add_hw =3D 0;
> 		if (event->attr.pinned) {
> 			perf_cgroup_event_disable(event, ctx);
> 			perf_event_set_state(event, =
PERF_EVENT_STATE_ERROR);
> +		} else {
> +			ctx->rotate_necessary =3D 1;
> +			perf_mux_hrtimer_restart(cpuctx);
> +			group_update_userpage(event);
> 		}
> -
> -		*can_add_hw =3D 0;
> -		ctx->rotate_necessary =3D 1;
> -		perf_mux_hrtimer_restart(cpuctx);
> 	}

Another optimisation that is possible in merge_sched_in is we can avoid =
calling "perf_mux_hrtimer_restart" multiple times if already =
rotate_necessary is set for that context. Even though =
"perf_mux_hrtimer_restart" will just return if hrtimer is already =
active, we could avoid the overhead of calling this function multiple =
times if there are many groups.=20

Something like below:

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 0c000cb01eeb..26eae79bd723 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -3731,8 +3731,10 @@ static int merge_sched_in(struct perf_event =
*event, void *data)
                }
=20
                *can_add_hw =3D 0;
-               ctx->rotate_necessary =3D 1;
-               perf_mux_hrtimer_restart(cpuctx);
+               if (!ctx->rotate_necessary) {
+                       ctx->rotate_necessary =3D 1;
+                       perf_mux_hrtimer_restart(cpuctx);
+               }
        }
=20
        return 0;


Thanks,
Athira Rajeev

>=20
> 	return 0;
> --=20
> 2.30.2
>=20