Letux-OS


How do you read log messages without it?
By doing what people have been done for decades prior to systemd: Using some other random-ass logger service like rsyslog or syslog-ng. Being a logger is not an init-system's job.

I am seriously baffled that you even had to ask about that.
 
Sorry, I've only been administering my own linux boxes for about twenty years now, and thus far I've always used prespun distributions which have always come with logging enabled, and have thus far always been satisfied that logs are getting generated and are accessible to me, and haven't felt the need to investigate further. I've never been trained as a systems administrator, I'm an applications coder by trade, and while bits and pieces of how the kernel and subsequent subsystems work have interested me, I've not gone into logging yet.
 
Short update about the USB-A interface of my Pandora: I dug out my ancient electrical knowledge, grabbed an old USB 2.0 cable, cut it in the middle and connected a voltage meter to the open end. The outcome is that the voltage meter shows 5V when running the SuperZaxxon Linux but 0V when running the letux Linux.
 
If the debugfs is not lying to me SuperZaxxon and Letux, both pulling the GPIO_164 up. This means that the switch (TPS2041B) for the USB_HOST_5V is on. This brings me to the conclusion that the power/voltage on the input of the switch is missing when running Letux. Reading the schematic further the problem might be the REGEN pin of the TWL4030 not enabling the TPS61029DRC to power the input of the TPS2041B
 
Below are some wild guesses. hns may, when he reads the text below, start laughing so loud that I will hear him (he is only 70 km away).

Sadly there is not much usage of the regen feature of the twl4030 in the OMAP3 dts files or I was not able to find it. I found only a comment in the file omap3-ldp.dts [1]. Looking at the patch that introduced the comment [2] the only stuff that looks twl4030 related are are two twl_gpios (7, 15). The (twl_)gpio 7 is mentioned together with LEDA in the dtsi file of the pandora. So the gpio left is 15. Does the device tree of the pandora just need another regulator been enabled via twl_gpio 15?

[1] https://git.kernel.org/pub/scm/linu...arch/arm/boot/dts/omap3-ldp.dts?h=v5.0.4#n192
[2] https://git.kernel.org/pub/scm/linu...s?id=bc10f310e2856f7bb80875fd7d8386836df4c85e
 
@hns

Hi today i would test the Pandora with a recent release of Letux.
Made a (micro)SD card with makesd but it stopped (black screen) at boot time...what is changed on the bootloader or i made something wrong?
I made the sdcard with:
makesd openpandora -r xfce
 
Hm. That is difficult to decide without looking at the serial console.
Did you press the right shoulder button and did choose the µSD card and then press the B button?
Screen should tell "executing script" for a moment and then screen goes dark. After ca. 10 seconds the left LEDs should start flickering/blinking.
 
Hm. That is difficult to decide without looking at the serial console.
Did you press the right shoulder button and did choose the µSD card and then press the B button?
Screen should tell "executing script" for a moment and then screen goes dark. After ca. 10 seconds the left LEDs should start flickering/blinking.

yes right..
message on screen say:
2129 bytes read in 20 ms (103.5 KiB/s)
## Exexcuting script at 82000000

and then screen goes dark (i waited around 20 min)...no led activity
 
Well, I have no idea at the moment. Do you have a 600MHz or an 1GHz Pandora? I only have a 600Mhz for testing so there may be a subtle bug in the 1GHz version which remained unnoticed.
BTW: WiFi is broken for a long time (since kernel letux-4.7-rc1) and I have planned to work on a fix...
 
I have also a 600Mhz (CC Pandora).
No problem and don't waste time to look at it if it take develop time for the Pyra...it's something that you could look at when you find some time.
 
WiFi is broken for a long time (since kernel letux-4.7-rc1) and I have planned to work on a fix...
It is working again! With letux-4.19 kernel or even letux-5.4-rc2.
I want to try to get the fixes upstream Linux so that they will also be available for distro kernels (e.g. Debian, although they won't have nub support).
Still open is controlling USB host port power. And bluetooth and audio.
@hns
Hi today i would test the Pandora with a recent release of Letux.
Made a (micro)SD card with makesd but it stopped (black screen) at boot time...what is changed on the bootloader or i made something wrong?
I made the sdcard with:
makesd openpandora -r xfce
I have baked a fresh SD and for me it did work. Hm. Did you update to latest makesd (makesd -u)?
 
It is working again! With letux-4.19 kernel or even letux-5.4-rc2.
I want to try to get the fixes upstream Linux so that they will also be available for distro kernels (e.g. Debian, although they won't have nub support).
Still open is controlling USB host port power. And bluetooth and audio.

I have baked a fresh SD and for me it did work. Hm. Did you update to latest makesd (makesd -u)?
Yes i was using makesd -u before ...but it didn't worked with my microSD card (inside an adapter)...i'll retry using a real SD card...but i have to wait as my linux box is out of order at the moment (the relatively old mechanical drive seems to refuse to boot...and i ordered a new SSD to replace).

Great work on fixing Wifi, next in line would be Audio ??
 
@hns
I've just replaced my faulty HardDrive with an SSD and now i'm back to business.
I have used a real SD Card this time but it didn't worked.
But this time i found an interesting thing:
If i first start booting the OS from the NAND and then rebooting with the one flashed on the SD card (both SD and micro via adapter) it works.
I don't know why.
Hope it helps found the cause.
 
Interesting. So now you have a method how to start and try out some LetuxOS flavours.

Maybe there is a different U-Boot in NAND flash which is incompatible with the LetuxOS startup?
My Pandora boots SuperZAXXON 1.76 and U-Boot console says:
Code:
Texas Instruments X-Loader 1.5.1 (Dec 10 2013 - 01:20:37)
OpenPandora System
OMAP3530-GP ES3.1
NAND: 2c:bc MT29F4G16ABBDA3W
Loading u-boot.bin from nand


U-Boot 2013.01.01 (Jul 20 2016 - 03:19:49)

OMAP3530-GP ES3.1, CPU-OPP2, L3-165MHz, Max CPU Clock 600 MHz
OMAP3 Pandora + LPDDR/NAND
I2C:   ready
DRAM:  512 MiB
NAND:  512 MiB
MMC:   OMAP SD/MMC: 0, OMAP SD/MMC: 1
*** Warning - bad CRC, using default environment
...
Loading file 'bootmenu.txt' to addr 0x82000000 with size 627 (0x00000273)...
Done
 
The Zaxxon version is shown on the boot screen when booting from NAND.
Well, the other info came from the console port (which needs some adapter hardware). I don't know if there is a different method to make it visible,
 
Back
Top