[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <c7c4864a-f3b7-6875-d229-c13b3808f7b3@electromag.com.au>
Date: Mon, 23 Sep 2019 09:33:51 +0800
From: Phil Reid <preid@...ctromag.com.au>
To: Pavel Machek <pavel@...x.de>, pavel@....cz
Cc: linux-kernel@...r.kernel.org, Moritz Fischer <mdf@...nel.org>,
Sasha Levin <sashal@...nel.org>
Subject: Re: [PATCH 4.19 32/79] fpga: altera-ps-spi: Fix getting of optional
confd gpio
On 22/09/2019 04:46, Pavel Machek wrote:
> Hi!
>
>> Currently the driver does not handle EPROBE_DEFER for the confd gpio.
>> Use devm_gpiod_get_optional() instead of devm_gpiod_get() and return
>> error codes from altera_ps_probe().
>
>> @@ -265,10 +265,13 @@ static int altera_ps_probe(struct spi_device *spi)
>> return PTR_ERR(conf->status);
>> }
>>
>> - conf->confd = devm_gpiod_get(&spi->dev, "confd", GPIOD_IN);
>> + conf->confd = devm_gpiod_get_optional(&spi->dev, "confd", GPIOD_IN);
>> if (IS_ERR(conf->confd)) {
>> - dev_warn(&spi->dev, "Not using confd gpio: %ld\n",
>> - PTR_ERR(conf->confd));
>> + dev_err(&spi->dev, "Failed to get confd gpio: %ld\n",
>> + PTR_ERR(conf->confd));
>> + return PTR_ERR(conf->confd);
>
> Will this result in repeated errors in dmesg in case of EPROBE_DEFER?
> We often avoid printing such messages in EPROBE_DEFER case.
Yes it will. I can submit a patch for that if required.
>
>> + } else if (!conf->confd) {
>> + dev_warn(&spi->dev, "Not using confd gpio");
>> }
>>
>> /* Register manager with unique name */
>
> Best regards,
> Pavel
>
--
Regards
Phil Reid
Powered by blists - more mailing lists