RTC issue on C4

GoogleGroups
1

I am looking at battery backup RTC issue since RTC is always reset to
1900-1-1 00:00:00 after power is off and
on again. This link http://e2e.ti.com/support/power_management/pmu/f/43/t/92666.aspx
gives me some clue. I checked C4 schematic. It seems TEST.RESET (T16)
is not grounded. If this is true, Why? how do I get it grounded? it
looks connected with other two pins.
This question may be right for Gerald Coley.

2

Do you have the battery installed on your board?

I just now managed to fix my kernel, that I custom compile for myself, so that the battery backed up clock can be written to with the hwclock command. Before I changed my kernel, the hwclock command operated but never changed the RTC.

I’m custom compiling so that I get one of the GPIO lines defined as a button for my usage, but here’s my diff if it is at all helpful.

diff linux-2.6.39/arch/arm/mach-omap2/board-omap3beagle.c linux-2.6.39.RFSAW/arch/arm/mach-omap2/board-omap3beagle.c
57a58,59

#define TWL4030_MSECURE_GPIO 22

518a521,526

{
.code = BTN_TOP,
.gpio = 130,
.desc = “GPIO-130-Button”,
.wakeup = 1,
},
679a688,692

3

Thanks! that is great job. I thought mSecure had been set somewhere.

4

This morning, I found hwclock is lost after one night's power-off.

5

Could it just be your battery? Could you confirm this with a good
battery and let us know?

Thanks,
Joel

6

That is a problem in the kernel.

See here:

http://www.spinics.net/lists/linux-omap/msg49596.html

Here is small patch for that:

http://www.spinics.net/lists/linux-omap/msg49620.html

But I've decided to use the following version, which also exports the GPIO so that the GPIO might be switched by root to disable modifications on the RTC.

7

The backup battery is 3V and the kernel is 2.6.32. But no joy.