Re: [PATCH] drm/panel: himax-hx8279: Initialize goa_{even,odd}_valid in hx8279_check_goa_config()

[AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>]

Il 23/04/25 15:50, Neil Armstrong ha scritto:
> On 23/04/2025 15:45, Neil Armstrong wrote:
>> On 22/04/2025 23:20, Nathan Chancellor wrote:
>>> Clang warns (or errors with CONFIG_WERROR=y):
>>>
>>>    drivers/gpu/drm/panel/panel-himax-hx8279.c:838:6: error: variable 
>>> 'goa_even_valid' is used uninitialized whenever 'if' condition is false [- 
>>> Werror,-Wsometimes-uninitialized]
>>>      838 |         if (num_zero == ARRAY_SIZE(desc->goa_even_timing))
>>>          |             ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>    drivers/gpu/drm/panel/panel-himax-hx8279.c:842:23: note: uninitialized use 
>>> occurs here
>>>      842 |         if (goa_odd_valid != goa_even_valid)
>>>          |                              ^~~~~~~~~~~~~~
>>>    drivers/gpu/drm/panel/panel-himax-hx8279.c:838:2: note: remove the 'if' if 
>>> its condition is always true
>>>      838 |         if (num_zero == ARRAY_SIZE(desc->goa_even_timing))
>>>          |         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>>>      839 |                 goa_even_valid = false;
>>>    drivers/gpu/drm/panel/panel-himax-hx8279.c:818:36: note: initialize the 
>>> variable 'goa_even_valid' to silence this warning
>>>      818 |         bool goa_odd_valid, goa_even_valid;
>>>          |                                           ^
>>>          |                                            = 0
>>>
>>> Even though only the even valid variable gets flagged, both valid
>>> variables appear to have the same issue of possibly being used
>>> uninitialized if the if statement initializing them to false is not
>>> taken.
>>>
>>> Initialize both values to true to clear up the warning and remove any
>>> possibility of encountering undefined behavior.
>>>
>>> Fixes: 38d42c261389 ("drm: panel: Add driver for Himax HX8279 DDIC panels")
>>> Signed-off-by: Nathan Chancellor <nathan@kernel.org>
>>> ---
>>>   drivers/gpu/drm/panel/panel-himax-hx8279.c | 2 +-
>>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>>
>>> diff --git a/drivers/gpu/drm/panel/panel-himax-hx8279.c b/drivers/gpu/drm/panel/ 
>>> panel-himax-hx8279.c
>>> index b48b350b62da..92f351e66c25 100644
>>> --- a/drivers/gpu/drm/panel/panel-himax-hx8279.c
>>> +++ b/drivers/gpu/drm/panel/panel-himax-hx8279.c
>>> @@ -815,7 +815,7 @@ static int hx8279_check_gmux_config(struct hx8279 *hx, 
>>> struct device *dev)
>>>   static int hx8279_check_goa_config(struct hx8279 *hx, struct device *dev)
>>>   {
>>>       const struct hx8279_panel_desc *desc = hx->desc;
>>> -    bool goa_odd_valid, goa_even_valid;
>>> +    bool goa_odd_valid = true, goa_even_valid = true;
>>>       int i, num_zero, num_clr = 0;
>>>       /* Up to 4 zero values is a valid configuration. Check them all. */
>>>
>>> ---
>>> base-commit: dcbd5dcc956e2331414fd7020b4655df08deeb87
>>> change-id: 20250422-panel-himax-hx8279-fix-sometimes-uninitialized-207354fb930c
>>>
>>> Best regards,
>>
>> Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
> 
> I'll wait a few days until AngeloGioacchino Del Regno reviews it to be sure
> it's the right fix.
> 
> Thanks,
> Neil

I would prefer

	if (num_zero == ARRAY_SIZE(desc->goa_odd_timing))
		goa_odd_valid = false;
	else
		goa_odd_valid = true;

	/* Up to 3 zeroes is a valid config. Check them all. */
	num_zero = 1;
	for (i = 0; i < ARRAY_SIZE(desc->goa_even_timing); i++) {
		if (desc->goa_even_timing[i])
			num_zero++;
	}

	if (num_zero == ARRAY_SIZE(desc->goa_even_timing))
		goa_even_valid = false;
	else
		goa_even_valid = true;

or the shorter form:

	goa_odd_valid = (num_zero != ARRAY_SIZE(desc->goa_odd_timing));

	/* Up to 3 zeroes is a valid config. Check them all. */
	num_zero = 1;
	for (i = 0; i < ARRAY_SIZE(desc->goa_even_timing); i++) {
		if (desc->goa_even_timing[i])
			num_zero++;
	}

	goa_even_valid = (num_zero != ARRAY_SIZE(desc->goa_even_timing));

...to avoid double initialization.

But anyway, the proposed solution also works just fine, as the sense is, as well
understood by Nathan, "if it's all zeroes something is wrong, otherwise it's ok".

Btw , I'm sorry about this - not sure how the code ended up being like this,
probably that last minute cleanup that I did.

Ultimately, whether to avoid the double init or not is Neil's call - so for this
commit you can get my

Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>

And sorry again about the (stupid) mistake.

Cheers,
Angelo