GPL Source Code

Theodore Borromeo

@WereCatf built a docker container buildroot for the lede project at borromeotlhs/omega2 on dockerhub.

I, probably incorrectly, chose the AR71xx as the target, but didn't see any other option. I figured a MIPS arch guess would be best. Also included objdump, motion and some other video streaming packages as I'm interested in using omega2 with webcams. It all built slowly with 'make -j1', and the resultant images are in that docker image (I think). Started pushing last night, and when I woke up this morning, my stupid laptop was restarting for updates, so I'm not sure it even finished pushing correctly

I'll post the dockerfile on github soon. Wife is mad at me for spending so much time yesterday getting docker up and running on Windows. . .

WereCatf

@Theodore-Borromeo said in GPL Source Code:

I, probably incorrectly, chose the AR71xx as the target, but didn't see any other option. I figured a MIPS arch guess would be best.

Why didn't you, you know, ask? There are things like e.g. endianness, which need to be correct -- you can't run big-endian binaries on a little-endian system! The correct profile is, in menuconfig, Mediatek Ralink MIPS -> MT7688 based boards -> Onion Omega2+, or manually, ramips omega2p.

Theodore Borromeo

@WereCatf I was just testing the ability for building to see if my Linux environment would build something, and I wanted to start from the omega 1 that I knew should have worked.

I'll fix that in v2 for the specific omega2

WereCatf

@Theodore-Borromeo said in GPL Source Code:

I'll fix that in v2 for the specific omega2

Okay, have fun with it and don't burn your build-machine down!

Theodore Borromeo

Building v2 now, will have it ready to push once done, and will post to the LEDE toolchain thread as well.

I'd love feedback once this is done as to what I could change about the docker image. Alternatively, folks can help @WereCatf out (not that he needs it) in making patches to his master branch, which is what I'm building off of. I had hoped that Onion,Inc. would build something like buildroot.org did to facilitate baking images to flash on your board once you got them. I think it'd aid greatly in this transparency discussion wrt GPL v2 and various other OSS licenses.

<3,
TJ

Larry Pinney

@administrators
@Kit-Bishop
@fossette

For those of you that can not run the command

' git show 5cd49bb067cadfeba6eb7f09982f76558101ad5d '

This will show the Onion Corporation :

By virtue of it's submission to OpenWrt has read and understands the terms of the GPL and those of OpenWrt.

Knowingly and willfully committed code and thus acknowledges the GPL as a valid license.

commit 5cd49bb067cadfeba6eb7f09982f76558101ad5d
Author: John Crispin john@openwrt.org
Date: Fri Jul 24 09:09:49 2015 +0000

ar71xx: add support for Onion Omega

This patch adds support for the Onion Omega.

https://onion.io/omega

Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
Acked-by: Boken Lin <bl@onion.io>
Tested-by: Jacky Huang <huangfangcheng@163.com>

SVN-Revision: 46458

###CUT##
Actual patch removed

committed code from :
target/linux/ar71xx/files/arch/mips/ath79/mach-onion-omega.c

Onion Omega board support

This program is free software; you can redistribute it and/or modify it
under the terms of the GNU General Public License version 2 as published by the Free Software Foundation.

###CUT###

Please push the code...as required.

Thanks for your contribution to Open Source Software...

WereCatf

@Lazar-Demin is trying to merge profiles for Omega2(+) to LEDE, but it seems they have not been listening to anything we've said here on the forums or taken a look at what we've done. Their DTS still contains the same issue with the reset-button that I mentioned several times already and that I've fixed in my DTS, for example.

Larry Pinney

@WereCatf
Yes ... I see we have a new branch.

It's very short of the complete source used.

However it is a start.

Larry Pinney

Here is my Device Tree Source :

"WE" should be able to fix a few things now

/dts-v1/;

#include "mt7628an.dtsi"

#include <dt-bindings/gpio/gpio.h>
#include <dt-bindings/input/input.h>

/ {
compatible = "onion,omega", "mediatek,mt7628an-soc";
model = "Onion Omega2";
chosen {

	bootargs = "console=ttyS0,115200";
};

gpio-leds {
	compatible = "gpio-leds";

	system {
		label = "omega2:amber:system";
		gpios = <&gpio1 12 GPIO_ACTIVE_LOW>;
	};
};

gpio-keys-polled {
	compatible = "gpio-keys-polled";
	#address-cells = <1>;
	#size-cells = <0>;
	poll-interval = <20>;

	reset {
		label = "reset";
		gpios = <&gpio1 6 GPIO_ACTIVE_HIGH>;
		linux,code = <KEY_RESTART>;
	};
};

};

&spi0 {
status = "okay";

pinctrl-names = "default";
pinctrl-0 = <&spi_pins>, <&spi_cs1_pins>;

m25p80@0 {
	#address-cells = <1>;
	#size-cells = <1>;
	compatible = "jedec,spi-nor";
	reg = <0>;
	linux,modalias = "m25p80";
	spi-max-frequency = <40000000>;
	m25p,chunked-io = <31>;

	partition@0 {
		label = "u-boot";
		reg = <0x0 0x30000>;
		read-only;
	};

	partition@30000 {
		label = "u-boot-env";
		reg = <0x30000 0x10000>;
	};

	factory: partition@40000 {
		label = "factory";
		reg = <0x40000 0x10000>;
		read-only;
	};

	partition@50000 {
		label = "firmware";
		reg = <0x50000 0x1fb0000>;
	};
};

spidev@1 {
	#address-cells = <1>;
	#size-cells = <1>;
	compatible = "linux,spidev";
	reg = <1>;
	spi-max-frequency = <40000000>;
};

};

&ethernet {
mtd-mac-address = <&factory 0x28>;
};

&gpio0 {
status = "okay";
};

&gpio1 {
status = "okay";
};

&gpio2 {
status = "okay";
};

&i2c {
status = "okay";
};

&pwm {
status = "okay";
};

&sdhci {
status = "okay";
mediatek,cd-high;
};

&uart1 {
status = "okay";
};

&uart2 {
status = "okay";
};

&wmac {
status = "okay";
ralink,mtd-eeprom = <&factory 0x4>;
};

&pinctrl {
state_default: pinctrl0 {

	gpio {
		ralink,group = "gpio", "i2c", "jtag", "perst", "pwm1", "refclk", "spis", "wdt", "wled_an", "wled_kn";
		ralink,function = "gpio";
	};
};

};

Larry Pinney

Not sure why this website mangles code ...

But if anybody wants it...I can send it another way

luz

@Larry-Pinney said in GPL Source Code:

@WereCatf
Yes ... I see we have a new branch.

It's very short of the complete source used.

However it is a start.

Yes, it's a start! Thanks @Lazar-Demin / greenbreakfast!

I hoped seeing the original Omega2 .dts would reveal why I can't get GPIO18/19 to work as GPIO in my LEDE build, while these work with standard Omega2 FW. However, no luck, the now posted .dts seems functionally identical to those figured out by @WereCatf and @Larry-Pinney (and my own).

From the voltage levels I see on these pins they are probably in Ethernet PHY mode, rather than GPIO, but I don't see why, and how to change that. omega2-ctrl does not seem to have an option for those pins.

Anyway, @onion please don't stop here! On my wishlist I see .config and the sources for omega2-ctrl in top positions

Luciano S.

@Larry-Pinney said in GPL Source Code:

Not sure why this website mangles code ...

This my help you:
https://community.onion.io/topic/707/markdown-how-to-start-with-it

WereCatf

I prettified my DTS-files up some and merged the reboot-fix by @Larry-Pinney , and now I am working on making a reasonably useable default-configuration that I can share soon.

Ghent The Slicer

Why all the fuss? I just managed to build the omega2 firmware from https://github.com/lede-project/source.git

Checkout the onion-omega2 branch
run:
scripts/feeds update -a
scripts/feeds install -a
make menuconfig

select CPU MIPS, Soc MT7688, Board Omega2 (or 2+ if you like)
run:
make V=99

enjoy

WereCatf

@Ghent-The-Slicer said in GPL Source Code:

Why all the fuss? I just managed to build the omega2 firmware from https://github.com/lede-project/source.git

Checkout the onion-omega2 branch

There is no such branch there. You may be referring to https://github.com/OnionIoT/source.git instead. Also, that branch appeared there less than two days ago, but this thread is quite a few days older than that already, so that's part of the "fuss." But sure, using your compiled image, try reboot... it won't work. Try the reset-button... it won't work.

WereCatf

Hm. The reboot-fix seems to have exposed another bug: the Omega2 now reboots fine, but doesn't shut down anymore! I don't know which one is better: to be able to reboot, or to be able to shutdown

Larry Pinney

With a device such as the Omega2...

Is the halt/shutdown command "really" useful?

Nice ..tend to agree...
But reboot seems to be used in practice much more often.

WereCatf

@Larry-Pinney That's kind of what I'm leaning towards. If you were to build some sort of a remote-sensor or whatever out of the Omega2, you'd probably want to be able to push an updated firmware, in case you have a bug somewhere, and be able to reboot to the new firmware without having to go and pull the plug to get it to reboot. But eh, I don't know.

Larry Pinney

Well...I think it's related to the bootloader....it always boots when powered on.

Nearest x86 equivalent could be "defined" in the BIOS....
Sorry we have NO BIOS today.

Ghent The Slicer

@WereCatf
I flashed the compiled image via USB drive:
copy the image to the usb drive and rename it to omega2.bin
reboot the omega2, holding the button, select 2 from the terminal, wait forever

It seems to boot fine except it is missing a large number of stuff - the web server for example. So you boot into serial console and that's about it.

Looks like some sort of build configuration issue to add the missing stuff.

Re: 'shutdown'. The omega2 with original firmware was not able to shutdown - it just reboots. Same with the compiled FW from the LEDE code. I have Linkit Smart 7688 - which has the same issue - no 'halt' - only reboot.