[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 8 Aug 2023 17:08:05 +0300
From: Andy Shevchenko <andriy.shevchenko@...ux.intel.com>
To: Marcus Folkesson <marcus.folkesson@...il.com>
Cc: Kent Gustavsson <kent@...oris.se>,
Jonathan Cameron <jic23@...nel.org>,
Lars-Peter Clausen <lars@...afoo.de>,
Rob Herring <robh+dt@...nel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
Conor Dooley <conor+dt@...nel.org>,
Cosmin Tanislav <demonsingur@...il.com>,
Arnd Bergmann <arnd@...db.de>,
ChiYuan Huang <cy_huang@...htek.com>,
Haibo Chen <haibo.chen@....com>,
Ramona Bolboaca <ramona.bolboaca@...log.com>,
Ibrahim Tilki <Ibrahim.Tilki@...log.com>,
ChiaEn Wu <chiaen_wu@...htek.com>,
William Breathitt Gray <william.gray@...aro.org>,
linux-iio@...r.kernel.org, devicetree@...r.kernel.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 2/4] iio: adc: mcp3911: simplify usage of spi->dev in
probe function
On Tue, Aug 08, 2023 at 01:04:30PM +0200, Marcus Folkesson wrote:
> Replace the usage of adc->spi->dev with spi->dev to make the code prettier.
Suggested-by: ?
> Signed-off-by: Marcus Folkesson <marcus.folkesson@...il.com>
...
> - adc->vref = devm_regulator_get_optional(&adc->spi->dev, "vref");
> + adc->vref = devm_regulator_get_optional(&spi->dev, "vref");
Why not
struct device *dev = &spi->dev;
and all the rest accordingly?
> if (IS_ERR(adc->vref)) {
> if (PTR_ERR(adc->vref) == -ENODEV) {
> adc->vref = NULL;
> } else {
> - dev_err(&adc->spi->dev,
> + dev_err(&spi->dev,
> "failed to get regulator (%ld)\n",
> PTR_ERR(adc->vref));
> return PTR_ERR(adc->vref);
Actually, you may first to switch to dev_err_probe() with the above introduced
struct device *dev = &spi->dev;
...
return dev_err_probe(dev, PTR_ERR(adc->vref),
"failed to get regulator\n",
and in the second patch do what you are doing here.
Will be much less changes and neater code at the end.
--
With Best Regards,
Andy Shevchenko
Powered by blists - more mailing lists