[PATCH v2 2/3] irqchip: Add the Sophgo SG2042 MSI interrupt controller

Chen Wang unicorn_wang at outlook.com
Fri Jan 10 17:29:53 PST 2025


On 2025/1/11 8:45, Inochi Amaoto wrote:
> On Mon, Dec 09, 2024 at 03:12:00PM +0800, Chen Wang wrote:
>> From: Chen Wang <unicorn_wang at outlook.com>
[......]
>> +
>> +struct sg2042_msi_data {
> suggestions: sg2042_msi_chipdata

ok

[......]

>> +static void sg2042_msi_irq_compose_msi_msg(struct irq_data *data,
>> +					   struct msi_msg *msg)
>> +{
>> +	struct sg2042_msi_data *priv = irq_data_get_irq_chip_data(data);
> It will be better to unify the variable name, it is confused for me.

ok, I will unify this from "priv" to "data" as other functions.

[......]

>> +	if (data->irq_first < SG2042_VECTOR_MIN ||
>> +	    (data->irq_first + data->num_irqs - 1) > SG2042_VECTOR_MAX) {
>> +		dev_err(&pdev->dev, "msi-ranges is incorrect!\n");
>> +		return -EINVAL;
>> +	}
> This check is fine, but I think it is kind of useless. Take the
> configuration for devicetree and process it as is, which also makes
> supporting new platform simple.

Sure, I will drop this check.

[......]

Thanks,

Chen





More information about the linux-riscv mailing list