On 21.8.2017 14:53, Alexandre Belloni wrote:

On 21/08/2017 at 14:47:20 +0200, Alexandre Belloni wrote:

quoted

On 21/08/2017 at 14:19:07 +0200, Michal Simek wrote:

quoted

On 21.8.2017 14:13, Alexandre Belloni wrote:

quoted

On 21/08/2017 at 12:50:34 +0200, Michal Simek wrote:

quoted

Hi,

On 21.8.2017 00:03, Alexandre Belloni wrote:

quoted

Hi Michal,

I've just send a patch to fix this issue (and avoid your other patch).

Could you test it? (I did test on an atmel platform)

Sorry it took so long!

not a problem but it looks like it is still just temporary solution. I
would expect that you can't unload module when this alarm is used. It if
it is not used then you should be able to remove it.

Yes, the alarmtimer handling needs to be made more flexible, especially
in the rtc device selection.

ok. What about to extend that commit message for the patch you sent to
explicitly say that this is temporary solution and what should be the
right fix?

Maybe you can also add link that patch which introduced this issue.
It was in that patch I sent.

You didn't pinpoint the correct patch. The patch introducing the issue
you were seeing (i.e. being able to remove the module) is
8bc0dafb5cf38a19484dfb16e2c6d29e85820046. Before this patch,
rtc_class_open() was called and this prevented the module from being
removed (it does try_module_get). For now, this is how the feature has
been implemented since 2011 and nobody cared enough to change that
behaviour.

To be clear, I'm not saying this should not be done. I'm just try to
explain that this would not be a fix but rather a feature improvement.

I am not quite sure I agree with this description. It doesn't mean that
none care about this from 2011 that this is correct behavior.
You know much more than I in this area but it seems to me quite normal
that if you have module and features which implement and you are not
using them that you can remove module.
It means your patch is a temporary solution for bug introduced in 2011
and found in 2016/2017.

Thanks,
Michal




-- 
You received this message because you are subscribed to "rtc-linux".
Membership options at http://groups.google.com/group/rtc-linux .
Please read http://groups.google.com/group/rtc-linux/web/checklist
before submitting a driver.
--- 
You received this message because you are subscribed to the Google Groups "rtc-linux" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rtc-linux+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.