OMAP5432 at home !

[Linux-SWAT]

When rebuilding the kernel, is it mandatory to append the DTB blob to the end of zImage ?

 

[aTc]

Not really, u-boot's bootz command can handle dtb files, the appending is more a hack for older bootloaders that don't have native support for it.




 

 

[Linux-SWAT]

Ok thanks.

I'm still fighting to compile a kernel on the igep.
I finally tried to compile their official kernel using gcc 4.9.3 natively, it just doesn't boot... I followed the strict instructions, nothing comes out after u-boot saying the kernel is loading.
I don't understand why it's supposed to work with a toolchain, and it doesn't with gcc.

 

[aTc]

Yeah, that's a bit of a problem with the 3.8 kernel, it just doesn't work when compiled with a more recent gcc.
I crosscompiled my working 3.8 igep kernel with gcc-linaro-arm-linux-gnueabihf-4.7

 

[Linux-SWAT]

I was able to recompile the 3.8.13 using Slackware 14.0 and gcc 4.7.1 .
Thanks.

 

[Linux-SWAT]

IGEP seems now to have "mainstream" support :
https://www.isee.biz/support/is-it-possible-to-get-an-updated-omap5-igep0050-dts-please#reply-8691

I'm currently compiling the 4.3 kernel, let's cross fingers !

 

[Linux-SWAT]

Not perfect yet, random hangs.
But at least it's a step in the good direction !!!

 

[Linux-SWAT]

On the EVM, is the little battery near boot switches supposed to keep the time ?

 

[hns]

On the EVM, is the little battery near boot switches supposed to keep the time ?

Well, not the battery is keeping the time - but it is intended to keep the RTC inside the Palmas running  What I don't know is if it is being charged. And the palmas needs a "rtc" node in the device tree.

 

[Linux-SWAT]

Each time I boot the device, the time is reset.

 

[hns]

Looks as if we have to add two lines to the device tree to make the backup battery being charged:

http://lxr.free-electrons.com/source/Documentation/devicetree/bindings/rtc/rtc-palmas.txt

 

[Linux-SWAT]

It's strange that it's still not implemented in the default dt.

 

[hns]

Indeed. Probably because nobody has submitted a patch for the omap5432 evm yet...

There is some rework by the OMAP maintainers so that 4.4-rc1 (coming in 2 weeks) will have a new shared omap5-common.dtsi.

This makes it easier to submit a proper patch. Although there should also be some palmas.dtsi which does not exist.

Again something on our long to-do list (not at highest priority).

 

[Linux-SWAT]

Yup, this is good news.

https://kernel.googlesource.com/pub/scm/linux/kernel/git/tmlind/linux-omap/

Any plan over heat management ?

For now I use a dumb script, maybe you have something better :

while true
do
 for t in $(sensors | grep temp | awk -F "+" '{ print $2 }' | awk -F "." '{ print $1 }'); do
  if [ $t -ge "90" ]; then
   echo "Too hot -" $(date | awk -F " " '{ print $4 }')
   cpufreq-set -u 1GHz
   echo "1GHz"
   sleep 30
  else
   cpufreq-set -u 1.5GHz
  fi
 done
sleep 2 
done

 

[TrashyMG]

Just a curious seeing hns is watching this thread, why was 500Mhz removed as a valid frequency in the dt? 

 

[hns]

Just a curious seeing hns is watching this thread, why was 500Mhz removed as a valid frequency in the dt? 

No idea. I even don't know if it was ever included... If you find an answer (some TI guys probably know), please share.

It could be possible to add it back. Notaz might know more about how to.

 

[notaz]

It was "descoped":

https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/arch/arm/boot/dts/omap5.dtsi?id=05e7d1a57a268b491d0e24952d219a73c62bd3df

 

[TrashyMG]

odd... shame 500Mhz actually still had a bit of performance.

 

[Linux-SWAT]

And it seems that 1.7GHz is also officially dropped.

 

[ible]

pardon my ignorance, but there was a way to run the CPU at 0.5 GHz?  (i assume that it would require less power, right?)  if we get proper 3D acceleration, that would still probably be plenty for certain games/activities...

since it's a kernel thing, that means you'd have to reboot to get to these other frequencies, right?