linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Lecopzer Chen <lecopzer.chen@mediatek.com>
To: <linux-kernel@vger.kernel.org>
Cc: <lecopzer.chen@mediatek.com>, <acme@kernel.org>,
	<akpm@linux-foundation.org>, <alexander.shishkin@linux.intel.com>,
	<catalin.marinas@arm.com>, <davem@davemloft.net>,
	<jolsa@redhat.com>, <jthierry@redhat.com>,
	<keescook@chromium.org>, <kernelfans@gmail.com>,
	<linux-arm-kernel@lists.infradead.org>,
	<linux-mediatek@lists.infradead.org>,
	<linux-perf-users@vger.kernel.org>, <mark.rutland@arm.com>,
	<masahiroy@kernel.org>, <matthias.bgg@gmail.com>,
	<maz@kernel.org>, <mcgrof@kernel.org>, <mingo@redhat.com>,
	<namhyung@kernel.org>, <nixiaoming@huawei.com>,
	<peterz@infradead.org>, <pmladek@suse.com>,
	<sparclinux@vger.kernel.org>, <sumit.garg@linaro.org>,
	<wangqing@vivo.com>, <will@kernel.org>, <yj.chiang@mediatek.com>
Subject: [PATCH v2 4/5] kernel/watchdog: Adapt the watchdog_hld interface for async model
Date: Mon, 7 Mar 2022 23:47:28 +0800	[thread overview]
Message-ID: <20220307154729.13477-5-lecopzer.chen@mediatek.com> (raw)
In-Reply-To: <20220307154729.13477-1-lecopzer.chen@mediatek.com>

When lockup_detector_init()->watchdog_nmi_probe(), PMU may be not ready
yet. E.g. on arm64, PMU is not ready until
device_initcall(armv8_pmu_driver_init).  And it is deeply integrated
with the driver model and cpuhp. Hence it is hard to push this
initialization before smp_init().

But it is easy to take an opposite approach by enabling watchdog_hld to
get the capability of PMU async.

The async model is achieved by expanding watchdog_nmi_probe() with
-EBUSY, and a re-initializing work_struct which waits on a wait_queue_head.

Co-developed-by: Pingfan Liu <kernelfans@gmail.com>
Signed-off-by: Pingfan Liu <kernelfans@gmail.com>
Signed-off-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
Suggested-by: Petr Mladek <pmladek@suse.com>
---
 include/linux/nmi.h |  3 +++
 kernel/watchdog.c   | 62 +++++++++++++++++++++++++++++++++++++++++++--
 2 files changed, 63 insertions(+), 2 deletions(-)

diff --git a/include/linux/nmi.h b/include/linux/nmi.h
index b7bcd63c36b4..cc7df31be9db 100644
--- a/include/linux/nmi.h
+++ b/include/linux/nmi.h
@@ -118,6 +118,9 @@ static inline int hardlockup_detector_perf_init(void) { return 0; }
 
 void watchdog_nmi_stop(void);
 void watchdog_nmi_start(void);
+
+extern bool lockup_detector_pending_init;
+extern struct wait_queue_head hld_detector_wait;
 int watchdog_nmi_probe(void);
 void watchdog_nmi_enable(unsigned int cpu);
 void watchdog_nmi_disable(unsigned int cpu);
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index b71d434cf648..49bdcaf5bd8f 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -103,7 +103,11 @@ void __weak watchdog_nmi_disable(unsigned int cpu)
 	hardlockup_detector_perf_disable();
 }
 
-/* Return 0, if a NMI watchdog is available. Error code otherwise */
+/*
+ * Arch specific API. Return 0, if a NMI watchdog is available. -EBUSY if not
+ * ready, and arch code should wake up hld_detector_wait when ready. Other
+ * negative value if not support.
+ */
 int __weak __init watchdog_nmi_probe(void)
 {
 	return hardlockup_detector_perf_init();
@@ -839,16 +843,70 @@ static void __init watchdog_sysctl_init(void)
 #define watchdog_sysctl_init() do { } while (0)
 #endif /* CONFIG_SYSCTL */
 
+static void lockup_detector_delay_init(struct work_struct *work);
+bool lockup_detector_pending_init __initdata;
+
+struct wait_queue_head hld_detector_wait __initdata =
+		__WAIT_QUEUE_HEAD_INITIALIZER(hld_detector_wait);
+
+static struct work_struct detector_work __initdata =
+		__WORK_INITIALIZER(detector_work, lockup_detector_delay_init);
+
+static void __init lockup_detector_delay_init(struct work_struct *work)
+{
+	int ret;
+
+	wait_event(hld_detector_wait,
+			lockup_detector_pending_init == false);
+
+	/*
+	 * Here, we know the PMU should be ready, so set pending to true to
+	 * inform watchdog_nmi_probe() that it shouldn't return -EBUSY again.
+	 */
+	lockup_detector_pending_init = true;
+	ret = watchdog_nmi_probe();
+	if (ret) {
+		pr_info("Delayed init of the lockup detector failed: %d\n", ret);
+		pr_info("Perf NMI watchdog permanently disabled\n");
+		return;
+	}
+
+	nmi_watchdog_available = true;
+	lockup_detector_setup();
+	lockup_detector_pending_init = false;
+}
+
+/* Ensure the check is called after the initialization of PMU driver */
+static int __init lockup_detector_check(void)
+{
+	if (!lockup_detector_pending_init)
+		return 0;
+
+	pr_info("Delayed init checking failed, retry for once.\n");
+	lockup_detector_pending_init = false;
+	wake_up(&hld_detector_wait);
+	return 0;
+}
+late_initcall_sync(lockup_detector_check);
+
 void __init lockup_detector_init(void)
 {
+	int ret;
+
 	if (tick_nohz_full_enabled())
 		pr_info("Disabling watchdog on nohz_full cores by default\n");
 
 	cpumask_copy(&watchdog_cpumask,
 		     housekeeping_cpumask(HK_FLAG_TIMER));
 
-	if (!watchdog_nmi_probe())
+	ret = watchdog_nmi_probe();
+	if (!ret)
 		nmi_watchdog_available = true;
+	else if (ret == -EBUSY) {
+		lockup_detector_pending_init = true;
+		queue_work_on(smp_processor_id(), system_wq, &detector_work);
+	}
+
 	lockup_detector_setup();
 	watchdog_sysctl_init();
 }
-- 
2.25.1


  parent reply	other threads:[~2022-03-07 15:48 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-07 15:47 [PATCH v2 0/5] Suppot hld based on Pseudo-NMI for arm64 Lecopzer Chen
2022-03-07 15:47 ` [PATCH v2 1/5] kernel/watchdog: remove WATCHDOG_DEFAULT Lecopzer Chen
2022-03-07 15:47 ` [PATCH v2 2/5] kernel/watchdog: change watchdog_nmi_enable() to void Lecopzer Chen
2022-03-07 15:47 ` [PATCH v2 3/5] kernel/watchdog_hld: Ensure CPU-bound context when creating hardlockup detector event Lecopzer Chen
2022-03-07 15:47 ` Lecopzer Chen [this message]
2022-03-18 10:40   ` [PATCH v2 4/5] kernel/watchdog: Adapt the watchdog_hld interface for async model Petr Mladek
2022-03-19  8:18     ` Lecopzer Chen
2022-03-21 17:37       ` Petr Mladek
2022-03-24 12:55         ` Lecopzer Chen
2022-03-07 15:47 ` [PATCH v2 5/5] arm64: Enable perf events based hard lockup detector Lecopzer Chen

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=20220307154729.13477-5-lecopzer.chen@mediatek.com \
    --to=lecopzer.chen@mediatek.com \
    --cc=acme@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=catalin.marinas@arm.com \
    --cc=davem@davemloft.net \
    --cc=jolsa@redhat.com \
    --cc=jthierry@redhat.com \
    --cc=keescook@chromium.org \
    --cc=kernelfans@gmail.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mediatek@lists.infradead.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=masahiroy@kernel.org \
    --cc=matthias.bgg@gmail.com \
    --cc=maz@kernel.org \
    --cc=mcgrof@kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=nixiaoming@huawei.com \
    --cc=peterz@infradead.org \
    --cc=pmladek@suse.com \
    --cc=sparclinux@vger.kernel.org \
    --cc=sumit.garg@linaro.org \
    --cc=wangqing@vivo.com \
    --cc=will@kernel.org \
    --cc=yj.chiang@mediatek.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).