[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ae880c97-e38d-0f01-f349-a427ab4afd41@loongson.cn>
Date: Mon, 19 Jun 2023 15:49:12 +0800
From: Yingkun Meng <mengyingkun@...ngson.cn>
To: Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
Harshit Mogalapalli <harshit.m.mogalapalli@...cle.com>,
broonie@...nel.org, lgirdwood@...il.com
Cc: linux-kernel@...r.kernel.org, alsa-devel@...a-project.org,
loongarch@...ts.linux.dev, loongson-kernel@...ts.loongnix.cn,
Dan Carpenter <dan.carpenter@...aro.org>,
Dan Carpenter <error27@...il.com>,
kernel-janitors@...r.kernel.org
Subject: Re: [PATCH v4 1/3] ASoC: Add support for Loongson I2S controller
On 2023/6/19 15:00, Krzysztof Kozlowski wrote:
> On 19/06/2023 03:45, Yingkun Meng wrote:
>>>>> + tx_data->irq = fwnode_irq_get_byname(fwnode, "tx");
>>>> Smatch detects that tx_data->irq and rx_data->irq are of type
>>>> u32(unsigned) so they can never be negative.
>>>>
>>>>> + if (tx_data->irq < 0) {
>>>> ^^^^^^^^ This can never be true.
>>>>
>>>> Should irq be of type 'int' instead?
>>>>
>>>>> + dev_err(&pdev->dev, "dma tx irq invalid\n");
>>>>> + return tx_data->irq;
>>>>> + }
>>>>> +
>>>>> + rx_data->irq = fwnode_irq_get_byname(fwnode, "rx");
>>>>> + if (rx_data->irq < 0) {
>>>> ^^^ Same problem here.
>>>>
>>>> Should irq
>>> Should 'irq' be of type int instead?
>>>
>>> As fwnode_irq_get_byname() returns a integer.
>>>
>> Yes, you are right. I will add a patch to fix the type of 'irq' to int.
> Run smatch and sparse on your code before posting. It would find such
> trivial mistakes.
Thanks, got it.
Thanks,
Yingkun
>
> Best regards,
> Krzysztof
Powered by blists - more mailing lists