On Thu, Nov 07, 2024 at 08:38:20PM +0100, Christophe JAILLET wrote: > Le 07/11/2024 à 20:18, Frank Li a écrit : > > Use devm_* and dev_err_probe() simplify probe function and remove > > vf610_adc_remove(). Change type of 'vref_uv' to int because > > regulator_get_voltage() return type is int. > > > > Reviewed-by: Haibo Chen <haibo.chen-3arQi8VN3Tc@xxxxxxxxxxxxxxxx> > > Signed-off-by: Frank Li <Frank.Li-3arQi8VN3Tc@xxxxxxxxxxxxxxxx> > > --- > > Change from v2 to v3 > > - change vref_uv to int from u32 to fix below warning > > | Reported-by: kernel test robot <lkp-ral2JQCrhuEAvxtiuMwx3w@xxxxxxxxxxxxxxxx> > > | Closes: https://lore.kernel.org/oe-kbuild-all/202411070633.NIrO7Ert-lkp-ral2JQCrhuEAvxtiuMwx3w@xxxxxxxxxxxxxxxx/ > > smatch warnings: > > drivers/iio/adc/vf610_adc.c:857 vf610_adc_probe() warn: unsigned 'info->vref_uv' is never less than zero. > > > > vim +857 drivers/iio/adc/vf610_adc.c > > > > Change from v1 to v2 > > - add Haibo's review tag > > --- > > drivers/iio/adc/vf610_adc.c | 79 ++++++++++--------------------------- > > 1 file changed, 20 insertions(+), 59 deletions(-) > > > > diff --git a/drivers/iio/adc/vf610_adc.c b/drivers/iio/adc/vf610_adc.c > > index 4d83c12975c53..a6a0ada8a102f 100644 > > --- a/drivers/iio/adc/vf610_adc.c > > +++ b/drivers/iio/adc/vf610_adc.c > > @@ -160,7 +160,7 @@ struct vf610_adc { > > /* lock to protect against multiple access to the device */ > > struct mutex lock; > > - u32 vref_uv; > > + int vref_uv; > > u32 value; > > struct regulator *vref; > > @@ -823,10 +823,8 @@ static int vf610_adc_probe(struct platform_device *pdev) > > int ret; > > indio_dev = devm_iio_device_alloc(&pdev->dev, sizeof(struct vf610_adc)); > > - if (!indio_dev) { > > - dev_err(&pdev->dev, "Failed allocating iio device\n"); > > - return -ENOMEM; > > - } > > + if (!indio_dev) > > + return dev_err_probe(&pdev->dev, -ENOMEM, "Failed allocating iio device\n"); > > info = iio_priv(indio_dev); > > info->dev = &pdev->dev; > > @@ -842,27 +840,22 @@ static int vf610_adc_probe(struct platform_device *pdev) > > ret = devm_request_irq(info->dev, irq, > > vf610_adc_isr, 0, > > dev_name(&pdev->dev), indio_dev); > > - if (ret < 0) { > > - dev_err(&pdev->dev, "failed requesting irq, irq = %d\n", irq); > > - return ret; > > - } > > + if (ret < 0) > > + dev_err_probe(&pdev->dev, ret, "failed requesting irq, irq = %d\n", irq); > > missing return? Yes, thank you find it. > > > - info->clk = devm_clk_get(&pdev->dev, "adc"); > > - if (IS_ERR(info->clk)) { > > - dev_err(&pdev->dev, "failed getting clock, err = %ld\n", > > - PTR_ERR(info->clk)); > > - return PTR_ERR(info->clk); > > - } > > + info->clk = devm_clk_get_enabled(&pdev->dev, "adc"); > > + if (IS_ERR(info->clk)) > > + return dev_err_probe(&pdev->dev, PTR_ERR(info->clk), > > + "failed getting clock, err = %ld\n", > > + PTR_ERR(info->clk)); > > No need to add an extra PTR_ERR(info->clk) yes, dev_err_probe() already print error code. So "err= %ld" can be removed also. > > > info->vref = devm_regulator_get(&pdev->dev, "vref"); > > With the change to devm_regulator_get_enable_read_voltage(), is it still > needed? Suspend function need vref to disable regulator. > > CJ > > > if (IS_ERR(info->vref)) > > return PTR_ERR(info->vref); > > - ret = regulator_enable(info->vref); > > - if (ret) > > - return ret; > > - > > - info->vref_uv = regulator_get_voltage(info->vref); > > + info->vref_uv = devm_regulator_get_enable_read_voltage(&pdev->dev, "vref"); > > + if (info->vref_uv < 0) > > + return info->vref_uv; > > device_property_read_u32_array(dev, "fsl,adck-max-frequency", info->max_adck_rate, 3); > > ... >
