From: Daniel Baluta <daniel.baluta@intel.com>
To: jic23@kernel.org
Cc: knaack.h@gmx.de, lars@metafoo.de, pmeerw@pmeerw.net,
linux-iio@vger.kernel.org, linux-kernel@vger.kernel.org,
wsa@the-dreams.de, linux-i2c@vger.kernel.org,
lucas.demarchi@intel.com, daniel.baluta@intel.com,
srinivas.pandruvada@linux.intel.com, ggao@invensense.com,
adi.reus@gmail.com, cmo@melexis.com, mwelling@ieee.org
Subject: [RFC PATCH 9/9] iio: imu: inv_mpu6050: Fix deadlock between i2c adapter lock and mpu lock
Date: Thu, 18 Feb 2016 17:53:14 +0200 [thread overview]
Message-ID: <1455810794-3188-10-git-send-email-daniel.baluta@intel.com> (raw)
In-Reply-To: <1455810794-3188-1-git-send-email-daniel.baluta@intel.com>
From: Adriana Reus <adriana.reus@intel.com>
This deadlock occurs if the accel/gyro and the sensor on the auxiliary
I2C (in my setup it's an ak8975) are working at the same time.
Scenario:
T1 T2
==== ====
inv_mpu6050_read_fifo aux sensor op (eg. ak8975_read_raw)
| |
mutex_lock(&indio_dev->mlock) i2c_transfer
| |
i2c transaction i2c adapter lock
| |
i2c adapter lock i2c_mux_master_xfer
|
inv_mpu6050_select_bypass
|
mutex_lock(&indio_dev->mlock)
When we operate on an mpu sensor the order of locking is mpu lock
followed by the i2c adapter lock. However, when we operate the auxiliary
sensor the order of locking is the other way around.
In order to avoid this enable the bypass mux bit once in the beginning
and remove the select/deselect_bypass functions.
This patch moves the bypass enabling in a separate function
that is called once at probe and removes the functionality from
inv_mpu_select/deselect_bypass.
Another advantage of this approach is that power-wise the mpu chip isn't
powered up at each auxiliary sensor i2c transaction so if only the
compass is used this would be more power efficient.
Signed-off-by: Adriana Reus <adriana.reus@intel.com>
Signed-off-by: Daniel Baluta <daniel.baluta@intel.com>
---
drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c | 88 ++++++-------------------------
1 file changed, 15 insertions(+), 73 deletions(-)
diff --git a/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c b/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c
index 71bdaa3..8065355 100644
--- a/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c
+++ b/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c
@@ -25,82 +25,25 @@ static const struct regmap_config inv_mpu_regmap_config = {
.val_bits = 8,
};
-/*
- * The i2c read/write needs to happen in unlocked mode. As the parent
- * adapter is common. If we use locked versions, it will fail as
- * the mux adapter will lock the parent i2c adapter, while calling
- * select/deselect functions.
- */
-static int inv_mpu6050_write_reg_unlocked(struct i2c_client *client,
- u8 reg, u8 d)
+static int inv_mpu6050_bypass_en(struct iio_dev *indio_dev)
{
- int ret;
- u8 buf[2] = {reg, d};
- struct i2c_msg msg[1] = {
- {
- .addr = client->addr,
- .flags = 0,
- .len = sizeof(buf),
- .buf = buf,
- }
- };
-
- ret = __i2c_transfer(client->adapter, msg, 1);
- if (ret != 1)
- return ret;
-
- return 0;
-}
-
-static int inv_mpu6050_select_bypass(struct i2c_adapter *adap, void *mux_priv,
- u32 chan_id)
-{
- struct i2c_client *client = mux_priv;
- struct iio_dev *indio_dev = dev_get_drvdata(&client->dev);
struct inv_mpu6050_state *st = iio_priv(indio_dev);
int ret = 0;
- /* Use the same mutex which was used everywhere to protect power-op */
- mutex_lock(&indio_dev->mlock);
- if (!st->powerup_count) {
- ret = inv_mpu6050_write_reg_unlocked(client,
- st->reg->pwr_mgmt_1, 0);
- if (ret)
- goto write_error;
+ ret = inv_mpu6050_set_power_itg(st, true);
+ if (ret)
+ return ret;
- msleep(INV_MPU6050_REG_UP_TIME);
- }
- if (!ret) {
- st->powerup_count++;
- ret = inv_mpu6050_write_reg_unlocked(client,
- st->reg->int_pin_cfg,
- INV_MPU6050_INT_PIN_CFG |
- INV_MPU6050_BIT_BYPASS_EN);
+ ret = regmap_write(st->map,
+ st->reg->int_pin_cfg,
+ INV_MPU6050_INT_PIN_CFG |
+ INV_MPU6050_BIT_BYPASS_EN);
+ if (ret) {
+ inv_mpu6050_set_power_itg(st, false);
+ return ret;
}
-write_error:
- mutex_unlock(&indio_dev->mlock);
- return ret;
-}
-
-static int inv_mpu6050_deselect_bypass(struct i2c_adapter *adap,
- void *mux_priv, u32 chan_id)
-{
- struct i2c_client *client = mux_priv;
- struct iio_dev *indio_dev = dev_get_drvdata(&client->dev);
- struct inv_mpu6050_state *st = iio_priv(indio_dev);
-
- mutex_lock(&indio_dev->mlock);
- /* It doesn't really mattter, if any of the calls fails */
- inv_mpu6050_write_reg_unlocked(client, st->reg->int_pin_cfg,
- INV_MPU6050_INT_PIN_CFG);
- st->powerup_count--;
- if (!st->powerup_count)
- inv_mpu6050_write_reg_unlocked(client, st->reg->pwr_mgmt_1,
- INV_MPU6050_BIT_SLEEP);
- mutex_unlock(&indio_dev->mlock);
-
- return 0;
+ return inv_mpu6050_set_power_itg(st, false);
}
/**
@@ -129,7 +72,8 @@ static int inv_mpu_probe(struct i2c_client *client,
return PTR_ERR(regmap);
}
- result = inv_mpu_core_probe(regmap, client->irq, name, NULL);
+ result = inv_mpu_core_probe(regmap, client->irq, name,
+ inv_mpu6050_bypass_en);
if (result < 0)
return result;
@@ -137,9 +81,7 @@ static int inv_mpu_probe(struct i2c_client *client,
st->mux_adapter = i2c_add_mux_adapter(client->adapter,
&client->dev,
client,
- 0, 0, 0,
- inv_mpu6050_select_bypass,
- inv_mpu6050_deselect_bypass);
+ 0, 0, 0, NULL, NULL);
if (!st->mux_adapter) {
result = -ENODEV;
goto out_unreg_device;
--
2.5.0
next prev parent reply other threads:[~2016-02-18 15:51 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-18 15:53 [RFC PATCH 0/9] iio: Fix ABBA deadlock in inv-mpu6050 Daniel Baluta
2016-02-18 15:53 ` [RFC PATCH 1/9] iio: imu: inv_mpu6050: Fix multiline comments style Daniel Baluta
2016-02-21 20:36 ` Jonathan Cameron
2016-02-18 15:53 ` [RFC PATCH 2/9] iio: imu: inv_mpu6050: Fix Yoda conditions Daniel Baluta
2016-02-19 9:09 ` Crt Mori
2016-02-21 20:38 ` Jonathan Cameron
2016-03-01 21:23 ` Wolfram Sang
2016-02-18 15:53 ` [RFC PATCH 3/9] iio: imu: inv_mpu6050: Fix newlines to make code easier to read Daniel Baluta
2016-02-21 20:38 ` Jonathan Cameron
2016-02-18 15:53 ` [RFC PATCH 4/9] iio: imu: inv_mpu6050: Remove unnecessary parentheses Daniel Baluta
2016-02-21 20:39 ` Jonathan Cameron
2016-02-18 15:53 ` [RFC PATCH 5/9] iio: imu: inv_mpu6050: Delete space before comma Daniel Baluta
2016-02-18 15:53 ` [RFC PATCH 6/9] iio: imu: inv_mpu6050: Fix code indent for if statement Daniel Baluta
2016-02-21 20:40 ` Jonathan Cameron
2016-02-18 15:53 ` [RFC PATCH 7/9] iio: imu: inv_mpu6050: Fix alignment with open parenthesis Daniel Baluta
2016-02-21 20:41 ` Jonathan Cameron
2016-02-21 20:59 ` Joe Perches
2016-02-21 21:08 ` Jonathan Cameron
2016-02-18 15:53 ` [RFC PATCH 8/9] i2c: i2c-mux: Allow for NULL select callback Daniel Baluta
2016-03-01 20:30 ` Wolfram Sang
2016-03-01 20:38 ` Daniel Baluta
2016-02-18 15:53 ` Daniel Baluta [this message]
2016-02-18 18:17 ` [RFC PATCH 9/9] iio: imu: inv_mpu6050: Fix deadlock between i2c adapter lock and mpu lock Srinivas Pandruvada
2016-02-19 20:17 ` Ge Gao
2016-03-01 20:50 ` Wolfram Sang
2016-03-02 16:33 ` Daniel Baluta
2016-03-02 17:06 ` Wolfram Sang
2016-02-21 23:17 ` [RFC PATCH 0/9] iio: Fix ABBA deadlock in inv-mpu6050 Wolfram Sang
2016-02-22 9:43 ` Daniel Baluta
2016-02-26 15:52 ` Daniel Baluta
2016-03-03 23:09 ` Peter Rosin
2016-03-04 10:20 ` Daniel Baluta
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=1455810794-3188-10-git-send-email-daniel.baluta@intel.com \
--to=daniel.baluta@intel.com \
--cc=adi.reus@gmail.com \
--cc=cmo@melexis.com \
--cc=ggao@invensense.com \
--cc=jic23@kernel.org \
--cc=knaack.h@gmx.de \
--cc=lars@metafoo.de \
--cc=linux-i2c@vger.kernel.org \
--cc=linux-iio@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=lucas.demarchi@intel.com \
--cc=mwelling@ieee.org \
--cc=pmeerw@pmeerw.net \
--cc=srinivas.pandruvada@linux.intel.com \
--cc=wsa@the-dreams.de \
/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).