[PATCH i-g-t v4 09/17] tests/xe_exec_sip: Add sanity-after-timeout test

[Manszewski, Christoph]

Hi Zbigniew,

On 27.08.2024 10:19, Zbigniew Kempczyński wrote:
> On Fri, Aug 23, 2024 at 08:22:14PM +0200, Christoph Manszewski wrote:
>> Add a subtest that checks if we are able to submit workloads after gpu
>> was reset due to hung job.
>>
>> Signed-off-by: Christoph Manszewski <christoph.manszewski at intel.com>
>> ---
>>   tests/intel/xe_exec_sip.c | 39 ++++++++++++++++++++++++++++++++-------
>>   1 file changed, 32 insertions(+), 7 deletions(-)
>>
>> diff --git a/tests/intel/xe_exec_sip.c b/tests/intel/xe_exec_sip.c
>> index ea1770cd6..5d57d2c78 100644
>> --- a/tests/intel/xe_exec_sip.c
>> +++ b/tests/intel/xe_exec_sip.c
>> @@ -31,6 +31,9 @@
>>   
>>   #define SHADER_CANARY 0x01010101
>>   
>> +#define SHADER_HANG 0
>> +#define SHADER_WRITE 1
>> +
> 
> Please introduce a new enum for this, it's more readable.
> 
>>   static struct intel_buf *
>>   create_fill_buf(int fd, int width, int height, uint8_t color)
>>   {
>> @@ -50,21 +53,32 @@ create_fill_buf(int fd, int width, int height, uint8_t color)
>>   	return buf;
>>   }
>>   
>> -static struct gpgpu_shader *get_shader(int fd)
>> +static struct gpgpu_shader *get_shader(int fd, const int shadertype)
>>   {
>>   	static struct gpgpu_shader *shader;
>>   
>>   	shader = gpgpu_shader_create(fd);
>>   	gpgpu_shader__write_dword(shader, SHADER_CANARY, 0);
>> +
>> +	switch (shadertype) {
>> +	case SHADER_HANG:
>> +		gpgpu_shader__label(shader, 0);
>> +		gpgpu_shader__nop(shader);
>> +		gpgpu_shader__jump(shader, 0);
>> +		break;
>> +	case SHADER_WRITE:
>> +		break;
>> +	}
>> +
>>   	gpgpu_shader__eot(shader);
>>   	return shader;
>>   }
>>   
>> -static uint32_t gpgpu_shader(int fd, struct intel_bb *ibb, unsigned int threads,
>> -			     unsigned int width, unsigned int height)
>> +static uint32_t gpgpu_shader(int fd, struct intel_bb *ibb, const int shadertype,
>> +			     unsigned int threads, unsigned int width, unsigned int height)
>>   {
>>   	struct intel_buf *buf = create_fill_buf(fd, width, height, COLOR_C4);
>> -	struct gpgpu_shader *shader = get_shader(fd);
>> +	struct gpgpu_shader *shader = get_shader(fd, shadertype);
>>   
>>   	gpgpu_shader_exec(ibb, buf, 1, threads, shader, NULL, 0, 0);
>>   	gpgpu_shader_destroy(shader);
>> @@ -125,8 +139,10 @@ xe_sysfs_get_job_timeout_ms(int fd, struct drm_xe_engine_class_instance *eci)
>>    * SUBTEST: sanity
>>    * Description: check basic shader with write operation
>>    *
>> + * SUBTEST: sanity-after-timeout
>> + * Description: check basic shader execution after job timeout
>>    */
>> -static void test_sip(struct drm_xe_engine_class_instance *eci, uint32_t flags)
>> +static void test_sip(int shader, struct drm_xe_engine_class_instance *eci, uint32_t flags)
>>   {
>>   	unsigned int threads = 512;
>>   	unsigned int height = max_t(threads, HEIGHT, threads * 2);
>> @@ -153,7 +169,7 @@ static void test_sip(struct drm_xe_engine_class_instance *eci, uint32_t flags)
>>   	ibb = intel_bb_create_with_context(fd, exec_queue_id, vm_id, NULL, 4096);
>>   
>>   	igt_nsec_elapsed(&ts);
>> -	handle = gpgpu_shader(fd, ibb, threads, width, height);
>> +	handle = gpgpu_shader(fd, ibb, shader, threads, width, height);
>>   
>>   	intel_bb_sync(ibb);
>>   	igt_assert_lt_u64(igt_nsec_elapsed(&ts), timeout);
>> @@ -186,7 +202,16 @@ igt_main
>>   		fd = drm_open_driver(DRIVER_XE);
>>   
>>   	test_render_and_compute("sanity", fd, eci)
>> -		test_sip(eci, 0);
>> +		test_sip(SHADER_WRITE, eci, 0);
>> +
>> +	test_render_and_compute("sanity-after-timeout", fd, eci) {
>> +		test_sip(SHADER_HANG, eci, 0);
> 
> It seems job will hang, but what will happen if it will be submitted
> on LR mode? Doesn't it stuck here forever? Do we support eudebugger
> on !LR mode?

Sorry I missed this one. I haven't really thought what would happen in 
lr mode since the test simply doesn't use it. We no longer intend to 
support non-lr mode in eudebug, but that goes only for contexts marked 
as debugable through the 'DRM_XE_EXEC_QUEUE_SET_PROPERTY_EUDEBUG' exec 
queue property which also isn't set here. So this subtest shouldn't be 
affected by eudebug hence I left it in 'xe_exec_sip'.

Thanks,
Christoph


> 
> --
> Zbigniew
> 
>> +
>> +		xe_for_each_engine(fd, eci)
>> +			if (eci->engine_class == DRM_XE_ENGINE_CLASS_RENDER ||
>> +			    eci->engine_class == DRM_XE_ENGINE_CLASS_COMPUTE)
>> +				test_sip(SHADER_WRITE, eci, 0);
>> +	}
>>   
>>   	igt_fixture
>>   		drm_close_driver(fd);
>> -- 
>> 2.34.1
>>