From: Benjamin Berg <benjamin@sipsolutions.net>
To: linux-pm@vger.kernel.org
Cc: Zhang Rui <rui.zhang@intel.com>,
Benjamin Berg <benjamin.berg@intel.com>,
Hans de Goede <hansg@kernel.org>
Subject: [PATCH v2] thermal: use a custom lock class for intel x86_pkg_temp
Date: Tue, 24 Jun 2025 15:24:06 +0200 [thread overview]
Message-ID: <20250624132406.1485407-1-benjamin@sipsolutions.net> (raw)
From: Benjamin Berg <benjamin.berg@intel.com>
The intel driver has code paths that will take the tz->lock while the
cpuhp_state-up lock is held. As the cpuhp_state-up lock is used in other
code paths, it may happen that lockdep detects possible deadlocks
through unrelated thermal zone devices.
Fix these false positives by using a separate lockdep class for the
x86_pkg_temp thermal device.
Reported-by: Hans de Goede <hansg@kernel.org>
Closes: https://lore.kernel.org/linux-pm/e9d7ef79-6a24-4515-aa35-d1f2357da798@kernel.org/
Signed-off-by: Benjamin Berg <benjamin.berg@intel.com>
---
Hi,
I believe that this should solve the lockdep warning that Hans was
seeing. That said, I have not tested it much.
v2:
- Fix function name
- Mark lock class variable static
Benjamin
---
drivers/thermal/intel/x86_pkg_temp_thermal.c | 2 ++
drivers/thermal/thermal_core.c | 7 +++++++
include/linux/thermal.h | 8 ++++++++
3 files changed, 17 insertions(+)
diff --git a/drivers/thermal/intel/x86_pkg_temp_thermal.c b/drivers/thermal/intel/x86_pkg_temp_thermal.c
index 3fc679b6f11b..15d3c904eaa3 100644
--- a/drivers/thermal/intel/x86_pkg_temp_thermal.c
+++ b/drivers/thermal/intel/x86_pkg_temp_thermal.c
@@ -310,6 +310,7 @@ static int pkg_temp_thermal_trips_init(int cpu, int tj_max,
static int pkg_temp_thermal_device_add(unsigned int cpu)
{
+ static struct lock_class_key x86_pkg_temp_class;
struct thermal_trip trips[MAX_NUMBER_OF_TRIPS] = { 0 };
int id = topology_logical_die_id(cpu);
u32 eax, ebx, ecx, edx;
@@ -349,6 +350,7 @@ static int pkg_temp_thermal_device_add(unsigned int cpu)
err = PTR_ERR(zonedev->tzone);
goto out_kfree_zonedev;
}
+ thermal_zone_device_set_lock_class(zonedev->tzone, &x86_pkg_temp_class);
err = thermal_zone_device_enable(zonedev->tzone);
if (err)
goto out_unregister_tz;
diff --git a/drivers/thermal/thermal_core.c b/drivers/thermal/thermal_core.c
index 17ca5c082643..ff5c2e01904a 100644
--- a/drivers/thermal/thermal_core.c
+++ b/drivers/thermal/thermal_core.c
@@ -1657,6 +1657,13 @@ struct thermal_zone_device *thermal_tripless_zone_device_register(
}
EXPORT_SYMBOL_GPL(thermal_tripless_zone_device_register);
+void thermal_zone_device_set_lock_class(struct thermal_zone_device *tz,
+ struct lock_class_key *lock_class)
+{
+ lockdep_set_class_and_name(&tz->lock, lock_class, tz->type);
+}
+EXPORT_SYMBOL_GPL(thermal_zone_device_set_lock_class);
+
void *thermal_zone_device_priv(struct thermal_zone_device *tzd)
{
return tzd->devdata;
diff --git a/include/linux/thermal.h b/include/linux/thermal.h
index 0b5ed6821080..3cb4cf60b66d 100644
--- a/include/linux/thermal.h
+++ b/include/linux/thermal.h
@@ -240,6 +240,9 @@ struct thermal_zone_device *thermal_tripless_zone_device_register(
const struct thermal_zone_device_ops *ops,
const struct thermal_zone_params *tzp);
+void thermal_zone_device_set_lock_class(struct thermal_zone_device *tz,
+ struct lock_class_key *lock_class);
+
void thermal_zone_device_unregister(struct thermal_zone_device *tz);
void *thermal_zone_device_priv(struct thermal_zone_device *tzd);
@@ -290,6 +293,11 @@ static inline struct thermal_zone_device *thermal_tripless_zone_device_register(
const struct thermal_zone_params *tzp)
{ return ERR_PTR(-ENODEV); }
+static inline void
+thermal_zone_device_set_lock_class(struct thermal_zone_device *tz,
+ struct lock_class_key *lock_class)
+{ }
+
static inline void thermal_zone_device_unregister(struct thermal_zone_device *tz)
{ }
--
2.50.0
next reply other threads:[~2025-06-24 13:24 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-06-24 13:24 Benjamin Berg [this message]
2025-06-25 9:44 ` [PATCH v2] thermal: use a custom lock class for intel x86_pkg_temp Hans de Goede
2025-06-25 10:15 ` Berg, Benjamin
2025-06-25 11:34 ` Hans de Goede
2025-06-25 13:23 ` Hans de Goede
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=20250624132406.1485407-1-benjamin@sipsolutions.net \
--to=benjamin@sipsolutions.net \
--cc=benjamin.berg@intel.com \
--cc=hansg@kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=rui.zhang@intel.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