@Luke-Lowrey The images are at http://repo.onion.io/omega2/images/
In CLI do:
cd /tmp
wget http://repo.onion.io/omega2/images/omega2p-v0.1.6-b137.bin
sysupgrade omega2p-v0.1.6-b137.bin
Wait for the LED to turn off, then restart your Omega2p and wait several minutes while the LED blinks.
It seems quite a few people have defective Omega2's, what with solder-bridges or poor solderings, I am attaching a few high-res close-ups here as a comparison, so you can compare yours to mine and hopefully spot any differences. My Omega2+ works quite fine, so there are pictures of a working, fully-functional device. These are quite high-resolution and yes, I noticed that they are a tad bit out-of-focus, but I can try to take more in-focus ones later on, if anyone really needs it.
@Oene-Bakker The binary and the sources for it on the Wiki are for Omega1, not Omega2. I modified the sources for the Omega2 and included a pre-compiled binary in the package, download it here: https://dl.dropboxusercontent.com/u/11811685/checkHumidity.tar.gz
@Mikael-Djurfeldt I am planning to work on my custom-image and I did think about writing a terse how-to, but that's gonna take a couple of days.
@Costas-Costas The segmentation fault - error is due to missing /dev/mem, not due to something in fast-gpio being wrong. You just have to wait for Onion-devs to fix their kernel.
I've never made a single PCB before, let alone used Kicad, but after tussling with it for a day I managed to do my own take on a mini-dock. I skipped on the USB-TTL, because I am going to be using an external one anyways, if needed, nor do I need those reset-buttons, but I did decide to include ESD-protection on the USB-port's data-lines. I also added a header with a couple of extra power-pins, either for powering the device from, or to power any external modules I may want to connect to it.
@Marcel-Neculae said in Fixing Fast-GPIO for Omega2/2+:
So my questions are:
- Is there a way to manually fix /dev/mem (installing a package or enabling a kernel module somehow), without recompiling a custom firmware. If yes, how?
No.
- Does your Fast-GPIO work? If yes, what's the full version that you run? (ex. 0.19 bXYZ)
0.1.6 b137 is the last version where it was working.
@jorgegarciadev I already knew that the partition for the U-Boot environment was empty and it was just using the defaults it was configured with, but I just learned that the kernel is configured so that it ignores any bootargs supplied by U-Boot -- it won't do you any good, even if I told you how to configure fw_printenv.
Basically, you're shit out of luck here, unless you go with a custom image.
@DrWheetos said in Omega2+ not starting up:
I'm the OP and have followed the RMA procedure to the letter but have had no response from Onion other than to close my support ticket saying it will be resolved by the community. I got the ticket reopened on 28th Dec and it's still open with no response so far.
That's quite terrible, to be honest. Pinging @administrators might help, but they should have fixed this from the get-go and not just closed the ticket.
@Theodore-Borromeo This issue has been around for fricking years and no one has fixed it "properly" and no one, who actually has the skills to do it, is willing to! But then no one is also willing to accept this patch, even temporarily, which comes to my opinion: it's idiocy. Getting things working should be more important than aesthetics, especially if the issue lingers around for years and years and nothing is ever done about it!
Yes, fixing things up in pinctrl would possibly be the "most correct" way of fixing things, but it's not happening. On the other hand, this is just a simple two-line patch, so it would be extremely easy to apply for now and remove one day in the future, if someone decided to fix things "correctly."
@luz said in Omega2 Cross Compiler and Kernel Sources:
(but [read his disclaimer!]
Everyone's referring to me as a "he" when it should be "she." I know it's irrelevant as to the topic, but it irks me anyways 
@Guest To be quite frank, formatting of a post does matter. The easier the post is to read, the easier it is for people to take part in the discussion -- I, for example, just can't be arsed to spend my time trying to read a jumbled mess, when I could read 5 easy-to-read posts in the same time and possibly provide some useful advice in them. If one isn't willing to spend a wee bit of effort in making a coherent post with good formatting, then why should someone else expend effort in helping, either?
Luciano S did come off as a tad rude, yes, but he also had a valid point. On the other hand, you are complaining about him burdening the site with links, and yet, here you are making new, not-particularly-useful threads of your own. Pot and kettle, IMHO.
@frederic-raducki You have to run opkg update to fetch the list of packages from the repos. This list is kept in RAM, so every time you reboot your device it's lost and you have to run the command again. So, run the command again, then try opkg install nano and it should work fine.
What firmware version are you running? The Onion-devs forgot to enable CONFIG_DEVMEM in the newer firmware. The last working one is 0.1.6 b137
I received my PCBs today and I see that I've made a mistake: the ESD-protection prevents the USB-host port from working right. Removing the resistor and capacitor (or not soldering them in the first place) allows it to work. Leaving those out just means a slightly less protected USB-port, otherwise the board works fine, so I'm not in a hurry to design a fix. I'll do it when I feel like it.
(The microUSB-port is missing, because Chinaman sent me wrong connectors -- they don't match the holes on the PCB. Just gotta wait for the right ones to arrive in the mail.)
Here is my take on a pinout-diagram. I specifically wanted to mark all the bootstrap-pins clearly, in the hopes that it'd be easier to avoid messing with them unnecessarily, and I wanted a higher-resolution diagram, that can be printed out on paper, if needed.
The SVG-file is, as usual, available on my github-repo ( https://github.com/WereCatf/SBC-pinouts )
I have compiled libsoc and its python2- and python3-bindings ( https://github.com/jackmitch/libsoc ) for the Omega2. libsoc allows you to read or write without specifying the register, so go nuts with it, if you please. A python-example is available at https://github.com/jackmitch/libsoc/blob/master/test/i2c_test.py
libsoc itself: https://dl.dropboxusercontent.com/u/11811685/omega2-stuff/libsoc_0.8.2-230117-1_mipsel_24kc.ipk
python2-bindings: https://dl.dropboxusercontent.com/u/11811685/omega2-stuff/python-libsoc_0.8.2-230117-1_mipsel_24kc.ipk
python3-bindings: https://dl.dropboxusercontent.com/u/11811685/omega2-stuff/python3-libsoc_0.8.2-230117-1_mipsel_24kc.ipk
EDIT: Thought to add the LEDE-package Makefile for libsoc that I made: https://dl.dropboxusercontent.com/u/11811685/omega2-stuff/lede-libsoc-Makefile -- this is only useful for those who wish to build their own LEDE-packages, so everyone else can just ignore this.
I have to agree with the others: this sounds very much like trying to force a square peg into a round hole. Pair a simple MCU with a SoC and you get the best of both worlds, or just go with a more capable MCU from the get-go. There's plenty of choice there, with my favourites being STM32L0-series for the low-end and ESP32 for the high-end.
@Daniel-Boese Something along the lines of should do:
#!/bin/sh
while true; do
datetime=$(date +%d-%m-%y--%H-%M)
mkdir -p $datetime
if [ $? -eq 0 ]; then
cd $datetime
wget http://whatever.url/some.file
cd ..
fi
sleep 1h
done
Note that it doesn't do any error-checking or anything. Add error-checking and whatnot, however it suits your needs.