Posts made by Kris Kelvin

looks like technically it is possible. https://labs.mediatek.com/en/chipset/MT7688 (datasheet, page 102)
Need to be implemented on kernel level.

looks like technically it is possible. https://labs.mediatek.com/en/chipset/MT7688 (page 102)
Need to be implemented on kernel level.

Re: Alternative C++ Code for GPIO Access (now with Interrupts and EXPLED)

I found that some pins, by some reason, staying inaccessible (**ERROR** Invalid pin number, GPIOAccess::isPinUsable returns true in new-gpio), despite of that I made them as "inputs" with fast-gpio. Is there any way to make them "free"?

fast-gpio works just fine. But I would like to utilize IRQ instead of permanent polling.

[Omega2p]

success! I solve the issue

This is what I have changed in Makefile (for both: libnewgpio and gpioirq)
TOOL_BIN_DIR=<uor path>Omega/source/staging_dir/toolchain-mipsel_24kc_gcc-5.5.0_musl/bin
CC=$(TOOL_BIN_DIR)/mipsel-openwrt-linux-g++

then just compile with make (libnewgpio and gpioirq consequently)

@luz said in GPIO Interrupts Omega2:

insmod gpio-irq

Hi @Zerog @luz, I've managed to compile and insmod the kernel module for Omega2. Thanks to your hints
however, I can't win the fight with the gpioirq (which is a part of this package https://github.com/KitBishop/Omega-GPIO-I2C-Arduino/tree/master/gpioirq).

i've been trying with "toolchain-mips_34kc_gcc-4.8-linaro_uClibc-0.9.33.2" recommended in PDF doc. successfully, but it doesn't work on Omega2 (giving: syntax error: unexpected "(" (expecting ")") ) , probably by architecture/binary difference. The original Omega2 toolchain can't link it (.o fileas are baked) and producing errors like:

./../../../mipsel-openwrt-linux-musl/bin/ld: ../libnewgpio/bin/static/libnewgpio.a(GPIOPin.o): compiled for a big endian system and target is little endian
./../../../mipsel-openwrt-linux-musl/bin/ld: ../libnewgpio/bin/static/libnewgpio.a(GPIOPin.o): endianness incompatible with that of the selected emulation
./../../../mipsel-openwrt-linux-musl/bin/ld: failed to merge target specific data of file ../libnewgpio/bin/static/libnewgpio.a(GPIOPin.o)

Would appreciate for any recommendations!

Hi colleagues,
I know about /etc/rc.button/reset . However, I can't find what is launching this script on each FW RESET button triggering.

The idea is to remove FW RESET at all and make it normal GPIO. I.e. ideally remove all (or maximum possible without recompiling the kernel unnecessary activity from the system regarding this pin. How I could do that?

Thank you for any relevant hint.

After updating to the last Omega2+ firmware reset and reboot commands stop working properly.
Commands cause no result until -f (Force, don't go through init) option used.

Thanks for any hints!

BTW, what means by "init" in this context (see line taken from --help above)?

@Chris-Stratton thank you for the answer! I just tried with the LED on GPIO1 and it booting and working just fine.
However, my attempt of connection to FW_RST was fall to the factory reset by some reason...
maybe you have any idea?

Permanent "socket connection lost" orange message in the bottom right corner. It's just impossible to write anything. That works normally once in a week early in the morning. All other time that demonstrating errors of connection on any attempt to write comment or ask anything.

What about to move community to some classical simple engine, which is not utilizing Web Sockets and not so depending on server/browser limitations?

@György-Farkas thank you. The only detail that still makes me doubt that the GPIO 6 should floating on boot. Obviously LED + resistor to GND will work as pull-down connection...

@ccs-hello Omega2+

@Chris-Stratton Following the schematics (https://github.com/OnionIoT/Onion-Hardware/blob/master/Schematics/Omega.pdf) GPIO 9 8 7 6 are floating by default, while hardware SPI (B51, B50, A54, B49) connected to internal FLASH MX25L1605DM2I. So it is logical to assume that GPIO 9 8 7 6 are used for software SPI emulation, which could be utilized as microSD interface (not clear from schematics) and in the same time as external SPI (for that developers, who not using microSD, maybe). Native SPI most probably staying aside anyway.

Сould you give a link to the source that provides comprehensive information on this?

@ccs-hello, thank you! Am I right understand that it means that I could safely use SPI MISO pin as GPIO for my needs without worrying that it has any impact on other system functionality?

And here is another questions: how to disable this activity? Id' like to use this pin as a LED indicator

Is there any chance to make omega2+ switch to the minimal energy consumption mode?
... and then, probably, wake-up with a button?

Hi colleagues, I just connected LED to SPI MISO and found that there are some activity. Led blinking.
Is my assumption (that microSD share interface with SPI) correct?

Many thanks!

@György-Farkas thanks

by the way, I found another more informative thread about VOUT here (for whom it may be interesting)
https://community.onion.io/topic/1524/would-someone-clarify-the-exact-function-of-these-omega2-pins/9

@Kris-Kelvin I just measure it. looks like there is no output voltage at all on my OMEGA 2+ VOUT. 0v.

Hi everybody.
That would be interesting to know what is a target purpose of this pin on OMEGA 2+? Would it be possible to drive this pin somehow? What is the maximum load that VOUT could withstand?

Is anybody know, where I could find that? I've seen this repo, unfortunately, there are no power dock schematics. Maybe somebody know, where I could find it? thanks

Yep. You certainly can connect D+ and D- directly to slave USB device. However, you still have to power slave device with +5v ( according to USB standard). So there are 2 options: a) power with step-up converter b) power with external +5v power supply.