Re: [PATCH] drm/i915/selftest: Add throttle reason diagnostics to RPS selftests

["Anirban, Sk" <sk.anirban@intel.com>]

Hi,

On 09-12-2025 12:46 pm, Krzysztof Karas wrote:
> Hi Sk Anirban,
>
> On 2025-12-09 at 11:26:17 +0530, Sk Anirban wrote:
>> Report GPU throttle reasons when RPS tests fail to reach expected
>> frequencies or power levels.
>>
>> Signed-off-by: Sk Anirban <sk.anirban@intel.com>
>> ---
>>   drivers/gpu/drm/i915/gt/selftest_rps.c | 17 +++++++++++++++++
>>   1 file changed, 17 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/i915/gt/selftest_rps.c b/drivers/gpu/drm/i915/gt/selftest_rps.c
>> index 73bc91c6ea07..01c671e00e61 100644
>> --- a/drivers/gpu/drm/i915/gt/selftest_rps.c
>> +++ b/drivers/gpu/drm/i915/gt/selftest_rps.c
>> @@ -1138,6 +1138,7 @@ int live_rps_power(void *arg)
>>   	struct intel_engine_cs *engine;
>>   	enum intel_engine_id id;
>>   	struct igt_spinner spin;
>> +	u32 throttle;
>>   	int err = 0;
>>   
>>   	/*
>> @@ -1216,6 +1217,13 @@ int live_rps_power(void *arg)
>>   		if (11 * min.power > 10 * max.power) {
>>   			pr_err("%s: did not conserve power when setting lower frequency!\n",
>>   			       engine->name);
>> +
>> +			throttle = intel_uncore_read(gt->uncore,
>> +						     intel_gt_perf_limit_reasons_reg(gt));
>> +
>> +			pr_warn("%s: GPU throttled with reasons 0x%08x\n",
>> +				engine->name, throttle & GT0_PERF_LIMIT_REASONS_MASK);
>> +
> This feels like spamming the system messages. We are on error
> path already and exiting with -EINVAL, so printing an
> unconditional warning here is excessive, I think.
>
> [...]
Got it. Based on past experience, most failures occur due to throttling. 
However, I can switch it to pr_info since we already print pr_err 
beforehand.

Thanks,

Anirban

>