[PATCH igt 5/9] msm/mapping: Add additional subtests

Kamil Konieczny kamil.konieczny at linux.intel.com
Tue Jul 1 16:17:24 UTC 2025


Hi Rob,
On 2025-06-30 at 11:08:59 -0700, Rob Clark wrote:
> From: Rob Clark <robdclark at chromium.org>

Fix 'From: ' e-mail, also please add to all subjects 'tests/' prefix:

[PATCH igt 5/9] tests/msm/msm_mapping: Add additional subtests

Please apply that also to other tests in series.

Btw, why not a little more desciription in subject?

[PATCH igt 5/9] tests/msm/msm_mapping: Test inaccessible buffers

Regards,
Kamil

> 
> Test a few additional buffers that userspace should not have access to.
> 
> Signed-off-by: Rob Clark <rob.clark at oss.qualcomm.com>
> ---
>  tests/msm/msm_mapping.c | 30 ++++++++++++++++++++++++++++++
>  1 file changed, 30 insertions(+)
> 
> diff --git a/tests/msm/msm_mapping.c b/tests/msm/msm_mapping.c
> index 5afbcd081fc7..846385bb5206 100644
> --- a/tests/msm/msm_mapping.c
> +++ b/tests/msm/msm_mapping.c
> @@ -253,6 +253,36 @@ igt_main
>  		do_mapping_test(pipe, "shadow", false);
>  	}
>  
> +	igt_describe("Test pwrup_reglist mapping, should be inaccessible");
> +	igt_subtest("pwrup_reglist") {
> +		do_mapping_test(pipe, "pwrup_reglist", true);
> +		do_mapping_test(pipe, "pwrup_reglist", false);
> +	}
> +
> +	igt_describe("Test memptrs mapping, should be inaccessible");
> +	igt_subtest("memptrs") {
> +		/*
> +		 * This test will fail on older GPUs without HW_APRIV, but
> +		 * there isn't a good way to test that from userspace, short
> +		 * of maintaining a giant table.  Probably just easier to
> +		 * list it in xfails or skips for those GPUs.
> +		 */
> +		do_mapping_test(pipe, "memptrs", true);
> +		do_mapping_test(pipe, "memptrs", false);
> +	}
> +
> +	igt_describe("Test 'preempt_record ring0' mapping, should be inaccessible");
> +	igt_subtest("preempt_record_ring0") {
> +		do_mapping_test(pipe, "preempt_record ring0", true);
> +		do_mapping_test(pipe, "preempt_record ring0", false);
> +	}
> +
> +	igt_describe("Test 'preempt_smmu_info ring0' mapping, should be inaccessible");
> +	igt_subtest("preempt_smmu_info_ring0") {
> +		do_mapping_test(pipe, "preempt_smmu_info ring0", true);
> +		do_mapping_test(pipe, "preempt_smmu_info ring0", false);
> +	}
> +
>  	igt_fixture {
>  		igt_msm_pipe_close(pipe);
>  		igt_msm_dev_close(dev);
> -- 
> 2.50.0
> 


More information about the igt-dev mailing list