From: y@motorola.com
To: Mark Brown <broonie@opensource.wolfsonmicro.com>,
Liam Girdwood <lrg@slimlogic.co.uk>
Cc: Lun Chang <L.chang@motorola.com>, <linux-kernel@vger.kernel.org>,
Jeffrey Carlyle <jeff.carlyle@motorola.com>
Subject: [PATCH v2] regulator: avoid deadlock when disabling regulator with supply
Date: Fri, 8 Oct 2010 14:48:22 -0500 [thread overview]
Message-ID: <1286567302-11298-1-git-send-email-y> (raw)
In-Reply-To: <20101008185739.GA6098@opensource.wolfsonmicro.com>
From: Jeffrey Carlyle <jeff.carlyle@motorola.com>
I have a regulator A that sets regulator B as its supply. When I call
set_supply to add B as the supply for A, regulator A gets added to the
supply_list for regulator B.
When I call regulator_disable(A), I end up with a call chain like this:
regulator_disable(A)
> mutex_lock(A)
> _regulator_disable(A)
>> _regulator_disable(B)
>>> _notifier_call_chain(B)
>>>> mutex_lock(A)
Which results in dead lock since we are trying to acquire the mutex lock
for regulator A which we already hold.
This patch addresses this issue by moving the call to disable regulator
B outside of the lock aquired inside the initial call to
regulator_disable.
This change also addresses the issue of not acquiring the mutex for
regulator B before calling _regulator_disable(B).
Signed-off-by: Jeffrey Carlyle <jeff.carlyle@motorola.com>
---
drivers/regulator/core.c | 35 ++++++++++++++++++++++++++---------
1 files changed, 26 insertions(+), 9 deletions(-)
diff --git a/drivers/regulator/core.c b/drivers/regulator/core.c
index cc8b337..2d215b5 100644
--- a/drivers/regulator/core.c
+++ b/drivers/regulator/core.c
@@ -63,7 +63,8 @@ struct regulator {
};
static int _regulator_is_enabled(struct regulator_dev *rdev);
-static int _regulator_disable(struct regulator_dev *rdev);
+static int _regulator_disable(struct regulator_dev *rdev,
+ struct regulator_dev **supply_rdev_ptr);
static int _regulator_get_voltage(struct regulator_dev *rdev);
static int _regulator_get_current_limit(struct regulator_dev *rdev);
static unsigned int _regulator_get_mode(struct regulator_dev *rdev);
@@ -1348,7 +1349,8 @@ int regulator_enable(struct regulator *regulator)
EXPORT_SYMBOL_GPL(regulator_enable);
/* locks held by regulator_disable() */
-static int _regulator_disable(struct regulator_dev *rdev)
+static int _regulator_disable(struct regulator_dev *rdev,
+ struct regulator_dev **supply_rdev_ptr)
{
int ret = 0;
@@ -1376,8 +1378,7 @@ static int _regulator_disable(struct regulator_dev *rdev)
}
/* decrease our supplies ref count and disable if required */
- if (rdev->supply)
- _regulator_disable(rdev->supply);
+ *supply_rdev_ptr = rdev->supply;
rdev->use_count = 0;
} else if (rdev->use_count > 1) {
@@ -1407,17 +1408,29 @@ static int _regulator_disable(struct regulator_dev *rdev)
int regulator_disable(struct regulator *regulator)
{
struct regulator_dev *rdev = regulator->rdev;
+ struct regulator_dev *supply_rdev = NULL;
int ret = 0;
mutex_lock(&rdev->mutex);
- ret = _regulator_disable(rdev);
+ ret = _regulator_disable(rdev, &supply_rdev);
mutex_unlock(&rdev->mutex);
+
+ /* decrease our supplies ref count and disable if required */
+ while (supply_rdev != NULL) {
+ rdev = supply_rdev;
+
+ mutex_lock(&rdev->mutex);
+ _regulator_disable(rdev, &supply_rdev);
+ mutex_unlock(&rdev->mutex);
+ }
+
return ret;
}
EXPORT_SYMBOL_GPL(regulator_disable);
/* locks held by regulator_force_disable() */
-static int _regulator_force_disable(struct regulator_dev *rdev)
+static int _regulator_force_disable(struct regulator_dev *rdev,
+ struct regulator_dev **supply_rdev_ptr)
{
int ret = 0;
@@ -1436,8 +1449,7 @@ static int _regulator_force_disable(struct regulator_dev *rdev)
}
/* decrease our supplies ref count and disable if required */
- if (rdev->supply)
- _regulator_disable(rdev->supply);
+ *supply_rdev_ptr = rdev->supply;
rdev->use_count = 0;
return ret;
@@ -1454,12 +1466,17 @@ static int _regulator_force_disable(struct regulator_dev *rdev)
*/
int regulator_force_disable(struct regulator *regulator)
{
+ struct regulator_dev *supply_rdev = NULL;
int ret;
mutex_lock(®ulator->rdev->mutex);
regulator->uA_load = 0;
- ret = _regulator_force_disable(regulator->rdev);
+ ret = _regulator_force_disable(regulator->rdev, &supply_rdev);
mutex_unlock(®ulator->rdev->mutex);
+
+ if (supply_rdev)
+ regulator_disable(get_device_regulator(rdev_get_dev(supply_rdev)));
+
return ret;
}
EXPORT_SYMBOL_GPL(regulator_force_disable);
--
1.6.3.3
next prev parent reply other threads:[~2010-10-08 19:48 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-01-04 18:16 [PATCH 0/3] Regulator updates for 2.6.34 Mark Brown
2010-01-04 18:17 ` [PATCH 1/3] regulator: Add notifier event on regulator disable Mark Brown
2010-10-08 13:27 ` Jeffrey Carlyle
2010-10-08 16:30 ` Mark Brown
2010-10-08 18:43 ` [PATCH] regulator: avoid deadlock when disabling regulator with supply Jeffrey Carlyle
2010-10-08 18:57 ` Mark Brown
2010-10-08 19:48 ` y [this message]
2010-10-08 19:49 ` [PATCH v2] " Jeffrey Carlyle
2010-10-08 21:10 ` Mark Brown
2010-10-10 10:14 ` Liam Girdwood
2010-01-04 18:17 ` [PATCH 2/3] regulator: Allow regulators to specify the time taken to ramp on enable Mark Brown
2010-01-04 18:17 ` [PATCH 3/3] regulator: Implement enable_time() for WM835x ISINKs Mark Brown
2010-01-04 20:35 ` [PATCH 0/3] Regulator updates for 2.6.34 Liam Girdwood
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=1286567302-11298-1-git-send-email-y \
--to=y@motorola.com \
--cc=L.chang@motorola.com \
--cc=broonie@opensource.wolfsonmicro.com \
--cc=jeff.carlyle@motorola.com \
--cc=linux-kernel@vger.kernel.org \
--cc=lrg@slimlogic.co.uk \
/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
all inboxes | Powered by JetHome®