Re: [PATCH] staging: gpib: simplify and fix get_data_lines

[Osama Abdelkader <osama.abdelkader@gmail.com>]

On 8/27/25 10:07 AM, Dan Carpenter wrote:
> On Wed, Aug 27, 2025 at 10:15:33AM +0300, Dan Carpenter wrote:
>> On Wed, Aug 27, 2025 at 12:05:02AM +0200, Osama Abdelkader wrote:
>>> The function `get_data_lines()` in gpib_bitbang.c currently reads 8
>>> GPIO descriptors individually and combines them into a byte.
>>> This has two issues:
>>>
>>>   * `gpiod_get_value()` returns an `int` which may be negative on
>>>     error. Assigning it directly into a `u8` may propagate unexpected
>>>     values. Masking ensures only the LSB is used.
>> Using the last bit in an error code is not really "error handling"...
>>
>> What you could do instead would be something like:
>>
>> 	int ret;
>>
>> 	for (i = 0; i < 8; i++) {
>> 		ret |= (gpiod_get_value(lines[i]) & 1) << i;
>> 		if (ret < 0) {
>> 			pr_err("something failed\n");
>> 			return -EINVAL;
> I meant to write "return 0;".  It's type u8.
>
> Also that doesn't work.  The masks and shift mess it up.
>
> 	u8 val = 0;
> 	int ret;
>
> 	for (i = 0; i < 8; i++) {
> 		ret = gpiod_get_value(lines[i]);
> 		if (ret < 0) {
> 			pr_err("something failed\n");
> 			continue;
> 		}
> 		val |= ret << i;
> 	}
>
> 	return ~val;

We can change the return type to int and propagate the error, so:

static int get_data_lines(u8 *out)

{

	int val, i;

	u8 ret = 0;

	struct gpio_desc *lines[8] = { D01, D02, D03, D04, D05, D06, D07, D08 };
	for (i = 0; i < 8; i++) { 		val = gpiod_get_value(lines[i]);

		if (val < 0)

			return val; // propagate error

		ret |= (val & 1) << i;

	}

	*out = ~ret; 	return 0;

}

Then in the caller:

u8 data;
if (!get_data_lines(&data))

	priv->rbuf[priv->count++] = data;

or we print the error here, What do you think?

>
> regards,
> dan carpenter
>