Re: [PATCH] iio: imu: inv_mpu6050:Adding judgment dev which is maybe a void pointer

[Guoqing Chi <chi962464zy@163.com>]

On Mon, 7 Jun 2021 16:20:06 +0300
Alexandru Ardelean <ardeleanalex@gmail.com> wrote:

> On Mon, Jun 7, 2021 at 2:39 PM Guoqing Chi <chi962464zy@163.com>
> wrote:
> >
> > From: Guoqing Chi <chiguoqing@yulong.com>
> >
> > ff40: 00000000 95fffa74 00000000 e07b2198 00000000 00000007
> > 81391000 e07b2198 ff60: 00000007 81391000 81267834 81391000
> > 0000015a 81200f00 00000006 00000006 ff80: 00000000 81200514
> > 00000000 80bfbd88 00000000 00000000 00000000 00000000 ffa0:
> > 00000000 80bfbd90 00000000 801010e8 00000000 00000000 00000000
> > 00000000 ffc0: 00000000 00000000 00000000 00000000 00000000
> > 00000000 00000000 00000000 ffe0: 00000000 00000000 00000000
> > 00000000 00000013 00000000 00000000 00000000 [<808f3228>]
> > (inv_mpu_i2c_aux_setup) from [<808f1c18>]
> > (inv_mpu_core_probe+0x638/0x878) [<808f1c18>] (inv_mpu_core_probe)
> > from [<808f3100>] (inv_mpu_probe+0xe4/0x1f0) [<808f3100>]
> > (inv_mpu_probe) from [<80801c88>] (i2c_device_probe+0x238/0x26c)
> > [<80801c88>] (i2c_device_probe) from [<8063520c>]
> > (really_probe+0x218/0x348) [<8063520c>] (really_probe) from
> > [<806354b4>] (driver_probe_device+0x60/0x164) [<806354b4>]
> > (driver_probe_device) from [<80635760>]
> > (device_driver_attach+0x58/0x60) [<80635760>]
> > (device_driver_attach) from [<806357e8>]
> > (__driver_attach+0x80/0xbc) [<806357e8>] (__driver_attach) from
> > [<80633560>] (bus_for_each_dev+0x74/0xb4) [<80633560>]
> > (bus_for_each_dev) from [<80634548>] (bus_add_driver+0x160/0x1e4)
> > [<80634548>] (bus_add_driver) from [<80636018>]
> > (driver_register+0x7c/0x114) [<80636018>] (driver_register) from
> > [<80801644>] (i2c_register_driver+0x3c/0x80) [<80801644>]
> > (i2c_register_driver) from [<8010277c>]
> > (do_one_initcall+0x54/0x1b8) [<8010277c>] (do_one_initcall) from
> > [<81200f00>] (kernel_init_freeable+0x144/0x1e0) [<81200f00>]
> > (kernel_init_freeable) from [<80bfbd90>] (kernel_init+0x8/0x10c)
> > [<80bfbd90>] (kernel_init) from [<801010e8>]
> > (ret_from_fork+0x14/0x2c)
> >
> > If we do not use aux or forget config child node at dts,it will be
> > in dump.
> >
> > Signed-off-by: Guoqing Chi <chiguoqing@yulong.com>
> > ---
> >  drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c | 5 ++++-
> >  1 file changed, 4 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c
> > b/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c index
> > 95f16951c8f4..b0f45042a25f 100644 ---
> > a/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c +++
> > b/drivers/iio/imu/inv_mpu6050/inv_mpu_i2c.c @@ -77,11 +77,14 @@
> > static int inv_mpu_i2c_aux_setup(struct iio_dev *indio_dev) }
> >
> >         /* enable i2c bypass when using i2c auxiliary bus */
> > -       if (inv_mpu_i2c_aux_bus(dev)) {
> > +       if (dev != NULL && inv_mpu_i2c_aux_bus(dev)) {  
> 
> One point is that this solves the issue for the I2C variant, but
> doesn't seem to handle the SPI variant of the driver/device.
> So, maybe this should be checked for NULL in inv_mpu_core_probe().
> 
> But, the 'dev' reference comes from this assignment 'struct device
> *dev = indio_dev->dev.parent;', which looks impossible given that the
> IIO device-object was created with devm_iio_device_alloc().
> So, 'indio_dev->dev.parent' can't be NULL, as far as I can tell.
> 
> On what version of the kernel has this been seen?

Hi Alexandru:
In kernel 5.4.70_2.3.0,and find the devm_iio_device_alloc() differ from
new version,The problem occurs apply the newest mpu6050 driver to
kernel 5.4.70,i will check it.

Thanks.

> 
> >                 ret = regmap_write(st->map, st->reg->int_pin_cfg,
> >                                    st->irq_mask |
> > INV_MPU6050_BIT_BYPASS_EN); if (ret)
> >                         return ret;
> > +       } else {
> > +               dev_err(&indio_dev->dev, "Failed to enable i2c
> > bypass.\n");
> > +               return -EFAULT;
> >         }
> >
> >         return 0;
> > --
> > 2.17.1
> >