[igt-dev] [PATCH i-g-t v3 2/2] tests/kms_concurrent: Plug possible memory leaks

[Kahola, Mika]

Pushed! 

Thanks for the comments and reviews.

Cheers,
Mika

-----Original Message-----
From: Lisovskiy, Stanislav <stanislav.lisovskiy at intel.com> 
Sent: Friday, March 27, 2020 10:02 AM
To: Kahola, Mika <mika.kahola at intel.com>
Cc: igt-dev at lists.freedesktop.org; juhapekka.heikkila at gmail.com
Subject: Re: [PATCH i-g-t v3 2/2] tests/kms_concurrent: Plug possible memory leaks

On Thu, Mar 26, 2020 at 03:53:22PM +0200, Mika Kahola wrote:
> Free dynamically allocated memory that has a potential to leak memory.
> 
> Signed-off-by: Mika Kahola <mika.kahola at intel.com>
> ---
>  tests/kms_concurrent.c | 8 ++++++++
>  1 file changed, 8 insertions(+)
> 

TBH dynamically allocating memory here seems to be overkill anyway. We already know for sure that max planes that we are going to have is no more than 8, We could just set limit of max planes to lets say even 10 and just do all this stuff in stack, thus avoiding those issues.

Reviewed-by: Stanislav Lisovskiy <stanislav.lisovskiy at intel.com>

> diff --git a/tests/kms_concurrent.c b/tests/kms_concurrent.c index 
> 1a7c12fc..1403e990 100644
> --- a/tests/kms_concurrent.c
> +++ b/tests/kms_concurrent.c
> @@ -193,6 +193,10 @@ prepare_planes(data_t *data, enum pipe pipe, int max_planes,
>  				    max_planes, output);
>  
>  	igt_plane_set_fb(data->plane[primary->index], 
> &data->fb[primary->index]);
> +
> +	free(x);
> +	free(y);
> +	free(size);
>  }
>  
>  static int test_bandwidth(data_t *data, enum pipe pipe, igt_output_t 
> *output) @@ -228,6 +232,10 @@ test_plane_position_with_output(data_t *data, enum pipe pipe, int max_planes,
>  	while (i < iterations || loop_forever) {
>  		prepare_planes(data, pipe, max_planes, output);
>  		igt_display_commit2(&data->display, COMMIT_ATOMIC);
> +
> +		for (int c = 0; c < max_planes; c++)
> +			igt_remove_fb(data->drm_fd, &data->fb[c]);
> +
>  		i++;
>  	}
>  }
> --
> 2.20.1
>