On 06/15/2011 09:25 PM, Bruce Ashfield wrote:

On 11-06-15 10:23 PM, Lu Jingdong wrote:

On 06/16/2011 01:33 AM, Darren Hart wrote:

On 06/15/2011 09:52 AM, Bruce Ashfield wrote:

On 06/15/11 12:48, Darren Hart wrote:

On 06/15/2011 06:31 AM, Bruce Ashfield wrote:

On 06/15/11 09:25, Koen Kooi wrote:

Op 15 jun 2011, om 15:21 heeft Bruce Ashfield het volgende
geschreven:

On 06/15/11 06:41, Jingdong Lu wrote:

From: Jingdong Lu<jingdong.lu@...>

Ensure msecure is mux'd to be able to set the RTC for
beagleboard. Fixes bug [YOCTO #767]

Jingdong,

The fix looks reasonable to me, Darren has been refreshing the
beagleboard support as well, so I've cc'd him here to see if we
have this change present in the work in progress trees.

I do not have something similar queued.

What tests did you run to ensure that the RTC worked on the
beagleboard ? There was some debate around whether or not you
needed an add-on card to get this functionality available ?

The signal "msecure" on TPS65950 connects to "msecure" on OMAP3530 and
setting/writing RTC register is controlled by
OMAP3530 "msecure". So we should pull the "msecure" high on OMAP3530 so
that we can set/write RTC register on TPS65950.

We test RTC on beagleboard by adding this patch. RTC can be set a
correct date and
it works well when power on.

You need to remove some resistors and add a battery, or add an
expansionboard like the zippy or zippy2, which have a DS1307 RTC
with battery.

Because there is no backup battery on beagleboard C4, RTC can not be
saved when
power off. If we want to keep RTC without power, we should remove R66
and install
a battery on board.

Ack'd. For me, this is reasonable. Having a working RTC even
if it doesn't survive power cycle is worth having. In particular,
since the patch is simple.

Agreed.

Thanks Koen,

This confirms what I thought (zippy*), so clearly this is either a
board with that setup, or runtime testing was .. unique. Either way,
more details and research is needed for this one.

I'll wait to hear upstream status on this. Given the requirements for
use, I wouldn't mind just closing 767 as will not fix as it can't work
with the beagleboard out of the box anyway.

I'm ok with that resolution .. for the bug. But if the patch
isn't already in another upstream tree, it's worth having
(if benign to the rest of the variants) and then we can get
it sent upstream from there.

I can change the bug's status if you like.

I want to hear back from Jingdong regarding this patch first. Is this
new code? If so, then yes, let's get it to mainline as Bruce suggested.

This patch can be found on linux-omap tree.
http://git.kernel.org/?p=linux/kernel/git/tmlind/linux-omap-2.6.git;a=commit;h=e2a346a2a054f702fd76f328ff747b9ad9264a4c

perfect.

arm: omap3: beagle: Ensure msecure is mux'd to be able to set the RTC
Without msecure beeing high it isn't possible to set (or start)
the RTC.

Tested with a BeagleBoard C4.

Signed-off-by: Alexander Holler <holler@...>
Signed-off-by: Tony Lindgren <tony@...>

So maybe I should recommit it by adding some upstream status and author.

This is key. We must have this in our patch commits. I can provide
examples if you aren't familiar and Mark Hatle just sent out some oe-core
examples of the same thing.

In this case you'd put something like:

commit e2a346a2a054f7 from
git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap-2.6.git

after the shortlog, and before the long log. Add your Integrated-by:
to the sign off section of the patch.

I will just pull the patch and add Jingdong as the Integrated-by.

--
Darren

Cheers,

Bruce

regards,

Jingdong

Cheers,

Bruce

--
Darren

Cheers,

Bruce

regards,

Koen

_______________________________________________ yocto mailing list
yocto@... https://lists.yoctoproject.org/listinfo/yocto

--
Darren Hart
Intel Open Source Technology Center
Yocto Project - Linux Kernel