diff mbox series

[3/5] kernel/watchdog_hld: Ensure CPU-bound context when creating hardlockup detector event

Message ID 20220212104349.14266-4-lecopzer.chen@mediatek.com (mailing list archive)
State New, archived
Headers show
Series Support hld based on Pseudo-NMI for arm64 | expand

Commit Message

Lecopzer Chen Feb. 12, 2022, 10:43 a.m. UTC 
From: Pingfan Liu <kernelfans@gmail.com>

from: Pingfan Liu <kernelfans@gmail.com>

hardlockup_detector_event_create() should create perf_event on the
current CPU. Preemption could not get disabled because
perf_event_create_kernel_counter() allocates memory. Instead,
the CPU locality is achieved by processing the code in a per-CPU
bound kthread.

Add a check to prevent mistakes when calling the code in another
code path.

Signed-off-by: Pingfan Liu <kernelfans@gmail.com>
Co-developed-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
Signed-off-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
---
 kernel/watchdog_hld.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

Comments

Petr Mladek Feb. 25, 2022, 1:15 p.m. UTC | #1 
On Sat 2022-02-12 18:43:47, Lecopzer Chen wrote:
> From: Pingfan Liu <kernelfans@gmail.com>
> 
> from: Pingfan Liu <kernelfans@gmail.com>
> 
> hardlockup_detector_event_create() should create perf_event on the
> current CPU. Preemption could not get disabled because
> perf_event_create_kernel_counter() allocates memory. Instead,
> the CPU locality is achieved by processing the code in a per-CPU
> bound kthread.
> 
> Add a check to prevent mistakes when calling the code in another
> code path.
> 
> Signed-off-by: Pingfan Liu <kernelfans@gmail.com>
> Co-developed-by: Lecopzer Chen <lecopzer.chen@mediatek.com>
> Signed-off-by: Lecopzer Chen <lecopzer.chen@mediatek.com>

Reviewed-by: Petr Mladek <pmladek@suse.com>

Best Regards,
Petr
diff mbox series

Patch

diff --git a/kernel/watchdog_hld.c b/kernel/watchdog_hld.c
index 247bf0b1582c..96b717205952 100644
--- a/kernel/watchdog_hld.c
+++ b/kernel/watchdog_hld.c
@@ -165,10 +165,16 @@  static void watchdog_overflow_callback(struct perf_event *event,
 
 static int hardlockup_detector_event_create(void)
 {
-	unsigned int cpu = smp_processor_id();
+	unsigned int cpu;
 	struct perf_event_attr *wd_attr;
 	struct perf_event *evt;
 
+	/*
+	 * Preemption is not disabled because memory will be allocated.
+	 * Ensure CPU-locality by calling this in per-CPU kthread.
+	 */
+	WARN_ON(!is_percpu_thread());
+	cpu = raw_smp_processor_id();
 	wd_attr = &wd_hw_attr;
 	wd_attr->sample_period = hw_nmi_get_sample_period(watchdog_thresh);