Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Date:   Mon, 10 Jul 2023 08:13:21 -0000
From:   "tip-bot2 for Namhyung Kim" <tip-bot2@linutronix.de>
Sender: tip-bot2@linutronix.de
Reply-to: linux-kernel@vger.kernel.org
To:     linux-tip-commits@vger.kernel.org
Subject: [tip: perf/urgent] perf/x86: Fix lockdep warning in
 for_each_sibling_event() on SPR
Cc:     Greg Thelen <gthelen@google.com>,
        Namhyung Kim <namhyung@kernel.org>,
        "Peter Zijlstra (Intel)" <peterz@infradead.org>,
        stable@vger.kernel.org, x86@kernel.org,
        linux-kernel@vger.kernel.org
In-Reply-To: <20230704181516.3293665-1-namhyung@kernel.org>
References: <20230704181516.3293665-1-namhyung@kernel.org>
MIME-Version: 1.0
Message-ID: <168897680188.404.3122821140126581927.tip-bot2@tip-bot2>
Robot-ID: <tip-bot2@linutronix.de>
Robot-Unsubscribe: Contact <mailto:tglx@linutronix.de> to get blacklisted from these emails
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Precedence: bulk

The following commit has been merged into the perf/urgent branch of tip:

Commit-ID:     27c68c216ee1f1b086e789a64486e6511e380b8a
Gitweb:        https://git.kernel.org/tip/27c68c216ee1f1b086e789a64486e6511e380b8a
Author:        Namhyung Kim <namhyung@kernel.org>
AuthorDate:    Tue, 04 Jul 2023 11:15:15 -07:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Mon, 10 Jul 2023 09:52:20 +02:00

perf/x86: Fix lockdep warning in for_each_sibling_event() on SPR

On SPR, the load latency event needs an auxiliary event in the same
group to work properly.  There's a check in intel_pmu_hw_config()
for this to iterate sibling events and find a mem-loads-aux event.

The for_each_sibling_event() has a lockdep assert to make sure if it
disabled hardirq or hold leader->ctx->mutex.  This works well if the
given event has a separate leader event since perf_try_init_event()
grabs the leader->ctx->mutex to protect the sibling list.  But it can
cause a problem when the event itself is a leader since the event is
not initialized yet and there's no ctx for the event.

Actually I got a lockdep warning when I run the below command on SPR,
but I guess it could be a NULL pointer dereference.

  $ perf record -d -e cpu/mem-loads/uP true

The code path to the warning is:

  sys_perf_event_open()
    perf_event_alloc()
      perf_init_event()
        perf_try_init_event()
          x86_pmu_event_init()
            hsw_hw_config()
              intel_pmu_hw_config()
                for_each_sibling_event()
                  lockdep_assert_event_ctx()

We don't need for_each_sibling_event() when it's a standalone event.
Let's return the error code directly.

Fixes: f3c0eba28704 ("perf: Add a few assertions")
Reported-by: Greg Thelen <gthelen@google.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/20230704181516.3293665-1-namhyung@kernel.org
---
 arch/x86/events/intel/core.c | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c
index a149faf..2a284ba 100644
--- a/arch/x86/events/intel/core.c
+++ b/arch/x86/events/intel/core.c
@@ -3993,6 +3993,13 @@ static int intel_pmu_hw_config(struct perf_event *event)
 		struct perf_event *leader = event->group_leader;
 		struct perf_event *sibling = NULL;
 
+		/*
+		 * When this memload event is also the first event (no group
+		 * exists yet), then there is no aux event before it.
+		 */
+		if (leader == event)
+			return -ENODATA;
+
 		if (!is_mem_loads_aux_event(leader)) {
 			for_each_sibling_event(sibling, leader) {
 				if (is_mem_loads_aux_event(sibling))