[PATCH 2/2 i-g-t] tests/amd_deadlock: add sdma negative test

[vitaly prosyak]

Both patches look good to me. I'll merge them after they pass our CI checks. 

I've posted them for CI verification since you've added several SDMA resets, and these tests are executed unconditionally in our CI pipeline.

Reviewed-by Vitaly Prosyak <vitaly.prosyak at amd.com>


On 2024-08-28 04:34, Jesse.zhang at amd.com wrote:
> To enhance the SDMA negative test,  add a few new tests.
>
> Signed-off-by: Jesse Zhang <jesse.zhang at amd.com>
> ---
>  tests/amdgpu/amd_deadlock.c | 28 ++++++++++++++++++++++++++++
>  1 file changed, 28 insertions(+)
>
> diff --git a/tests/amdgpu/amd_deadlock.c b/tests/amdgpu/amd_deadlock.c
> index 96da49002..5dd3bf9f1 100644
> --- a/tests/amdgpu/amd_deadlock.c
> +++ b/tests/amdgpu/amd_deadlock.c
> @@ -127,6 +127,34 @@ igt_main
>  		}
>  	}
>  
> +	igt_describe("Test-GPU-reset-by-sdma-badop-with-jobs");
> +	igt_subtest_with_dynamic("amdgpu-deadlock-sdma-badop-test") {
> +		if (arr_cap[AMD_IP_DMA]) {
> +			igt_dynamic_f("amdgpu-deadlock-sdma-badop-test")
> +			bad_access_helper(device, CMD_STREAM_EXEC_INVALID_OPCODE,
> +					AMDGPU_HW_IP_DMA);
> +		}
> +	}
> +
> +
> +	igt_describe("Test-GPU-reset-by-sdma-bad-addr-with-jobs");
> +	igt_subtest_with_dynamic("amdgpu-deadlock-sdma-bad-addr-test") {
> +		if (arr_cap[AMD_IP_DMA]) {
> +			igt_dynamic_f("amdgpu-deadlock-sdma-bad-addr-test")
> +			bad_access_helper(device, CMD_STREAM_TRANS_BAD_MEM_ADDRESS,
> +					AMDGPU_HW_IP_DMA);
> +		}
> +	}
> +
> +	igt_describe("Test-GPU-reset-by-sdma-bad-length-with-jobs");
> +	igt_subtest_with_dynamic("amdgpu-deadlock-sdma-bad-length-test") {
> +		if (arr_cap[AMD_IP_DMA]) {
> +			igt_dynamic_f("amdgpu-deadlock-sdma-bad-length-test")
> +			bad_access_helper(device, CMD_STREAM_TRANS_BAD_MEM_ADDRESS,
> +					AMDGPU_HW_IP_DMA);
> +		}
> +	}
> +
>  	igt_fixture {
>  		amdgpu_device_deinitialize(device);
>  		drm_close_driver(fd);