[Gta04-owner] 3.16-rc5

Dr. H. Nikolaus Schaller hns at goldelico.com
Mon Jul 21 07:49:57 CEST 2014


Hi Neil,

Am 21.07.2014 um 00:55 schrieb Neil Jerram <neil at ossau.homelinux.net>:

> "Dr. H. Nikolaus Schaller" <hns at goldelico.com> writes:
> 
>> Another note: HDQ is also not working.
> 
> I don't really understand this at all, but I read the relevant issues
> and then googled a bit, and came across
> http://www.kernelhub.org/?msg=21497&p=2.  There doesn't appear to be any
> equivalent logic to that in our 3.16-rc5 source, so I naively tried to
> write an equivalent change, giving the attached patch.
> 
> <0001-OMAP-hdq-software-supervised-control-of-HDQ_ICLK.patch>
> After booting with that patch, I noticed that
> /sys/class/power_supply/twl4030_battery/uevent indicates "Charging",
> with a current of about 603mA, after I've unplugged and replugged the
> USB cable.  And, the capacity value is now gradually increasing.
> 
> Is that significant?  (Unfortunately I didn't do a similar experiment
> before trying this patch.)
> 
> Also are there other things that I should look at, to try to determine
> if HDQ is behaving any differently from before?

Well, what I know is that HDQ works really fine without any patch,
if we disable the WiFi/BT logic in the device tree.

So I suspect that we don’t have to patch HDQ at all - just find out what
goes wrong in WiFi/BT. Maybe some imbalanced IRQ disabling or
something which stops the HDQ low level stuff from working.

And we didn’t need to patch the clock frameworks for the non-DT kernels
used up to 3.12. So I am not sure if we are even trying the right thing with
the patch you have suggested.

Regarding the charging indication it should be independent of HDQ
because the twl_battery driver runs independently.

BR,
Nikolaus



More information about the Gta04-owner mailing list