Re: [PATCH 2/5] media: i2c: mt9v111: fix incorrect type for ret

[Qianfeng Rong]

在 2025/8/27 20:58, Jacopo Mondi 写道:
> [You don't often get email from jacopo.mondi@xxxxxxxxxxxxxxxx. Learn why this is important at https://aka.ms/LearnAboutSenderIdentification ]
>
> Hi Qianfeng
>
> On Wed, Aug 27, 2025 at 08:39:10PM +0800, Qianfeng Rong wrote:
> > Change "ret" from unsigned int to int type in mt9v111_calc_frame_rate()
> > to store negative error codes or zero returned by __mt9v111_hw_reset()
> > and other functions.
> >
> > Storing the negative error codes in unsigned type, doesn't cause an issue
> > at runtime but it's ugly as pants.
> >
> > No effect on runtime.
> well, I'm not sure that's true.
>
> The code goes as
>
>          ret = __mt9v111_hw_reset(mt9v111);
>          if (ret == -EINVAL)
>                  ret = __mt9v111_sw_reset(mt9v111);
>          if (ret)
>                  return ret;
>
> And if ret is unsigned the condition ret == -EINVAL will never be met.
>
> I guess this actually fixes a bug, doesn't it ?
> You can add:
>
> Cc: stable@xxxxxxxxxxxxxxx
> Fixes: aab7ed1c3927 ("media: i2c: Add driver for Aptina MT9V111")
> Reviewed-by: Jacopo Mondi <jacopo.mondi@xxxxxxxxxxxxxxxx>
>
> Thanks
>    j

I have written a test program on the arm64 platform:

unsigned int ret = -ENOMEM;

if (ret == -ENOMEM)
    pr_info("unsigned int ret(%u) == -ENOMEM\n", ret);
else
    pr_info("unsigned int ret(%u) != -ENOMEM\n", ret);

Output log is: unsigned int ret(4294967284) == -ENOMEM

I suspect that -ENOMEM is forcibly converted to an unsigned type during the
comparison, but I am not sure if this behavior is consistent across all
platforms and compilers. Therefore, I agree that your suggestion and will
submit the v2 version.

Best regards,
Qianfeng