Re: linux-next: Tree for Jun 13 (drivers/soc/qcom/Kconfig)

[Randy Dunlap <rdunlap@infradead.org>]

On 6/13/19 11:27 AM, Rafael J. Wysocki wrote:
> On Thursday, June 13, 2019 6:12:43 PM CEST Randy Dunlap wrote:
>>  This is a multi-part message in MIME format.
>> --------------D4992319A3EA7FFA5EF515C6
>> Content-Type: text/plain; charset=windows-1252
>> Content-Transfer-Encoding: 8bit
>>
>> On 6/13/19 2:30 AM, Stephen Rothwell wrote:
>>> Hi all,
>>>
>>> Changes since 20190612:
>>>
>>
>>
>> on x86_64:
>>
>> ../drivers/base/power/domain.c: In function �genpd_queue_power_off_work�:
>> ../drivers/base/power/domain.c:485:13: error: �pm_wq� undeclared (first use in this function)
>>   queue_work(pm_wq, &genpd->power_off_work);
>>              ^
>> ../drivers/base/power/domain.c:485:13: note: each undeclared identifier is reported only once for each function it appears in
>> ../drivers/base/power/domain.c: In function �genpd_dev_pm_qos_notifier�:
>> ../drivers/base/power/domain.c:675:25: error: �struct dev_pm_info� has no member named �ignore_children�
>>    if (!dev || dev->power.ignore_children)
>>                          ^
>> ../drivers/base/power/domain.c: In function �rtpm_status_str�:
>> ../drivers/base/power/domain.c:2754:16: error: �struct dev_pm_info� has no member named �runtime_error�
>>   if (dev->power.runtime_error)
>>                 ^
>> ../drivers/base/power/domain.c:2756:21: error: �struct dev_pm_info� has no member named �disable_depth�
>>   else if (dev->power.disable_depth)
>>                      ^
>> ../drivers/base/power/domain.c:2758:21: error: �struct dev_pm_info� has no member named �runtime_status�
>>   else if (dev->power.runtime_status < ARRAY_SIZE(status_lookup))
>>                      ^
>> ../drivers/base/power/domain.c:2759:31: error: �struct dev_pm_info� has no member named �runtime_status�
>>    p = status_lookup[dev->power.runtime_status];
>>                                ^
>>
>>
>> Full randconfig file is attached.
> 
> PM_GENERIC_DOMAINS, which is set, depend on PM, which is not set.

Thanks for that.  (I changed the subject to indicate the errant entity.)

> Looks like an impossible configuration to me.

It's what happens when a driver selects PM_GENERIC_DOMAINS when PM is not set/enabled.

From drivers/soc/qcom/Kconfig:

config QCOM_AOSS_QMP
	tristate "Qualcomm AOSS Driver"
	depends on ARCH_QCOM || COMPILE_TEST
	depends on MAILBOX
	depends on COMMON_CLK
	select PM_GENERIC_DOMAINS

In my test case, COMPILE_TEST=y.  ARCH_QCOM is not set.

And kconfig does warn:

WARNING: unmet direct dependencies detected for PM_GENERIC_DOMAINS
  Depends on [n]: PM [=n]
  Selected by [m]:
  - QCOM_AOSS_QMP [=m] && (ARCH_QCOM || COMPILE_TEST [=y]) && MAILBOX [=y] && COMMON_CLK [=y]


Who is responsible for that Kconfig segment?
Ah, Bjorn.  Cc-ed.

-- 
~Randy