[igt-dev] [PATCH i-g-t v2] tests/kms_big_fb: Use igt_assert on bufmgr

[Ville Syrjälä]

On Wed, Jul 22, 2020 at 07:17:58PM +0530, Pankaj Bharadiya wrote:
> NULL bufmgr pointer leads to SIGSEGV. Assert on NULL bufmgr so that
> correct failure reason will get dumped in the stack trace.
> 
> Signed-off-by: Pankaj Bharadiya <pankaj.laxminarayan.bharadiya at intel.com>
> ---
> * changes since v1:
>  - Make the error message more helpful [Petri]
> 
>  tests/kms_big_fb.c | 5 +++++
>  1 file changed, 5 insertions(+)
> 
> diff --git a/tests/kms_big_fb.c b/tests/kms_big_fb.c
> index a754b2997..48caa5cd1 100644
> --- a/tests/kms_big_fb.c
> +++ b/tests/kms_big_fb.c
> @@ -607,6 +607,7 @@ igt_main
>  {
>  	igt_fixture {
>  		drmModeResPtr res;
> +		const struct intel_device_info *info;
>  
>  		data.drm_fd = drm_open_driver_master(DRIVER_INTEL);
>  
> @@ -649,6 +650,10 @@ igt_main
>  			data.render_copy = igt_get_render_copyfunc(data.devid);
>  
>  		data.bufmgr = drm_intel_bufmgr_gem_init(data.drm_fd, 4096);
> +		info = intel_get_device_info(data.devid);
> +		igt_assert_f(data.bufmgr, "Update libdrm for %s support\n",
> +			     info->gen ? info->codename : "<unknown chipset>");
> +

How many individual tests are you going to have to patch like this?
Would probably make more sense to plop this into some wrapper.

>  		data.batch = intel_batchbuffer_alloc(data.bufmgr, data.devid);
>  	}
>  
> -- 
> 2.23.0
> 
> _______________________________________________
> igt-dev mailing list
> igt-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/igt-dev

-- 
Ville Syrjälä
Intel