Questions regarding the wakeup interrupt

Holger Freyther zecke at openmoko.org
Thu Mar 27 15:34:31 CET 2008


On Thursday 27 March 2008 06:43:38 matt_hsu wrote:
> 		Both of SMS and incoming calls can issue an interrupt to host CPU.
> Btw, here is the information about power management which posted on the
> kernel list:
>
> 		http://wiki.openmoko.org/wiki/Overview_on_power_management_design

Hey Matt,

the above wiki page does not say much. It does not answer the questions I have 
asked and also the test results "OK" does not say much for incoming call/sms.

What I miss from the tests:

	- The the setup is not described. Was gsmd used. Was minicom used? Which 
kernel version, which hardware revision, which firmware revision was used to 
do the test? This is like physics one describes the setup of the experiment.

	- How was it tested? Was the device put to sleep with echo "mem" 
to /sys/power/state

	- Then the device was called /sent a SMS to and it woke up?

	- Where is the log file of the modem communication. It would be nice to have 
an annotated version, where you annotate where the device was suspend and 
when it was woken up.



Reasoning of experiments/tests:
	One wants to verify a hypothesis.
	You create a setup (a test) and then you do the testing.
	You do the test and keep a log
	With software you redo the test after fixing the software

	While developing ideally each of these things will be a regression test, 
sometimes this is not feasible.


Main Reason for experiment/tests:
	Other people can verify the results and people start to trust the results of 
an experiment/test.


Back to the point:
	My modem is waking me up immediately, I have described my test in the initial 
mail (I didn't name the version numbers). The result doesn't match the result 
on the wiki page and I want to find out why.

z.

PS: Allen's test plans looked good on a first sight :)



More information about the openmoko-devel mailing list