[PATCH] data: klog.json: lower the level of the failure when the _S1 and _S2 not implemented (LP: #1279714)

IvanHu ivan.hu at canonical.com
Thu Feb 13 15:29:47 UTC 2014


It think the errors are from the drivers/acpi/sleep.c

static void acpi_sleep_suspend_setup(void)
{
	int i;

	for (i = ACPI_STATE_S1; i < ACPI_STATE_S4; i++) {
		acpi_status status;
		u8 type_a, type_b;

		status = acpi_get_sleep_type_data(i, &type_a, &type_b);
		if (ACPI_SUCCESS(status)) {
			sleep_states[i] = 1;
		}
	}

	suspend_set_ops(old_suspend_ordering ?
		&acpi_suspend_ops_old : &acpi_suspend_ops);
}
would like to settup all sleep states data
and the acpica(hwxface.c) print out that the _S1 and _S2 not found.

Not sure if we can say it is an error from kernel. Since ACPI spec does 
define _S1 and _S2, but most BIOS don't implement them.
I think it isn't even an error, I'll sent resend the patch to set 
LOG_LEVEL_INFO, and add some advice like
"Failed to find _S1 namespace object that contains the register values 
for the sleep state. This means that OSPM don't know how to enter S1 
sleep state, the S1 sleep state of the system isn't supported."

Sounds good? any comments?

Cheers,
Ivan

On 02/13/2014 09:21 PM, alex.hung at canonical.com wrote:
> I think they are kernel errors, not BIOS errors as they are never declared in ASL.
>
> I sent a patch but not sure what is the status right now.
>
> Sent from my iPad
>
>> On Feb 13, 2014, at 8:10 PM, Colin Ian King <colin.king at canonical.com> wrote:
>>
>>> On 13/02/14 09:57, Ivan Hu wrote:
>>> Many machines don't implement the sleep state objects _S1 and _S2 on their firmware
>>> due to these two sleep states are seldom used.
>>> And it will cause the fwts klog test high failure as below:
>>>
>>> klog: HIGH Kernel message: [ 0.291430] ACPI Exception: AE_NOT_FOUND, While evaluating Sleep State [\_S1_] (20130517/hwxface-571)
>>> klog: HIGH Kernel message: [ 0.291435] ACPI Exception: AE_NOT_FOUND, While evaluating Sleep State [\_S2_] (20130517/hwxface-571)
>>>
>>> Signed-off-by: Ivan Hu <ivan.hu at canonical.com>
>>> ---
>>> data/klog.json | 14 ++++++++++++++
>>> 1 file changed, 14 insertions(+)
>>>
>>> diff --git a/data/klog.json b/data/klog.json
>>> index da08962..ae447e8 100644
>>> --- a/data/klog.json
>>> +++ b/data/klog.json
>>> @@ -1053,6 +1053,20 @@
>>>    },
>>>    {
>>>      "compare_mode": "string",
>>> +    "log_level": "LOG_LEVEL_LOW",
>>> +    "pattern": "While evaluating Sleep State [\\_S1_]",
>>> +    "advice": "Failed to evaluate _Sx namespace object that contains the register values for the sleep state.",
>>> +    "label": "KlogAcpiSleepStateEvalFailed"
>>> +  },
>>> +  {
>>> +    "compare_mode": "string",
>>> +    "log_level": "LOG_LEVEL_LOW",
>>> +    "pattern": "While evaluating Sleep State [\\_S2_]",
>>> +    "advice": "Failed to evaluate _Sx namespace object that contains the register values for the sleep state.",
>>> +    "label": "KlogAcpiSleepStateEvalFailed"
>>> +  },
>>> +  {
>>> +    "compare_mode": "string",
>>>      "log_level": "LOG_LEVEL_HIGH",
>>>      "pattern": "While evaluating Sleep State",
>>>      "advice": "Failed to evaluate _Sx namespace object that contains the register values for the sleep state.",
>>
>> If possible, can we provide more advice on what the implications are on
>> this error. E.g. do we lose functionality or can these be ignored?
>>
>> Colin
>>
>> --
>> fwts-devel mailing list
>> fwts-devel at lists.ubuntu.com
>> Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/fwts-devel
>




More information about the fwts-devel mailing list