Enable OLED support for Teensy 3.2/LC

[zvecr]

Description

Another situation where the ChibiOS hal does a poor job of abstraction, where I2Cv1 has a different config structure.

Note: Teensy LC do not have enough memory to run a 128x32 OLED in the codebases current state.

Types of Changes

• Core
• Bugfix
• New feature
• Enhancement/optimization
• Keyboard (addition or update)
• Keymap/layout/userspace (addition or update)
• Documentation

Issues Fixed or Closed by This PR

• Fixes ARM I2C driver not implemented for KINETIS MCU's (i.e. Teensy 3.2/3.5/3.6/LC) #7574

Checklist

• My code follows the code style of this project.
• My change requires a change to the documentation.
• I have updated the documentation accordingly.
• I have read the CONTRIBUTING document.
• I have added tests to cover my changes.
• I have tested the changes and verified that they work and don't break anything (as well as I can manage).

[stale]

Thank you for your contribution!
This pull request has been automatically marked as stale because it has not had activity in the last 45 days. It will be closed in 30 days if no further activity occurs. Please feel free to give a status update now, or re-open when it's ready.
For maintainers: Please label with awaiting review, breaking_change, in progress, or on hold to prevent the issue from being re-flagged.

[laughingrice]

Not sure if it helps, I merged the changes into master and managed to compile both my code and handwired/onekey on the teensy 3.2 with the i2c_scanner keymap, but no i2c device is found (error -1).

I will be happy for any pointers/examples/missing patches. Not sure if the problem is the I2C code or if it's using non-standard I2C pins

[zvecr]

Just rebased and retested.

Default pins are configured to be 16 and 17 on this image,

With those, i2c scanner works fine.

[laughingrice]

I must be doing something wrong then.
I ran this sanity check / test using arduino and got "I2C device found at address 0x69"

#include <Wire.h>

void ScanI2C()
{
    int nDevices = 0;

    for(byte address = 1 ; address < 127 ; address++)
    {
        Wire.beginTransmission(address);
        byte error = Wire.endTransmission();

        if (!error)
        {
            Serial.print("I2C device found at address 0x");
            Serial.println(address,HEX);
            nDevices++;
        }
    }

    if (nDevices == 0)
        Serial.println("No I2C devices found\n");
}

void setup()
{
    Serial.begin(38400); // USB is always 12 Mbit/sec
    while (!Serial);

    Wire.begin();
    Wire.setSDA(17);
    Wire.setSCL(16);

    ScanI2C();
}

void loop() {}

I then tested the said branch:

git clone https://github.com/zvecr/qmk_firmware.git
cd qmk_firmware
git checkout feature/teensy_oled
make git-submodule

In keyboards/handwired/onekey/rules.mk changed the following

DEFAULT_FOLDER = handwired/onekey/teensy_32
LAYOUTS = i2c_scanner

ran qmk compile -kb handwired/onekey -km i2c_scanner, flashed .build/handwired_onekey_teensy_32_i2c_scanner.hex and got

Scanning...
No I2C devices found

[zvecr]

As a FYI, you don't need to modify the rules.mk, just run qmk compile -kb handwired/onekey/teensy_32 -km i2c_scanner.

I assume there must be a difference in the I2C arduino implementation, and the device i'm using for testing responds differently.

qmk_firmware/keyboards/handwired/onekey/keymaps/i2c_scanner/keymap.c

Lines 13 to 17 in 2f15cb2

	i2c_start(address);
	// except on ChibiOS where the only way is do do "something"
	uint8_t data = 0;
	return i2c_readReg(address, 0, &data, sizeof(data), TIMEOUT);

thus the "best effort" build warning.

[St0RM53]

@zvecr @drashna Anything i can test please let me know. My keyboard with teensy 3.2 has already oled connected to it via i2c that runs a demo if i flash it with an arduino based program, so hardware is functional. Just need you to point me to which code changes to re-compile and test.

[drashna]

From github itself, you should be able to run this:

git fetch origin refs/pull/7591/head
git checkout -b zvecr/feature/teensy_oled FETCH_HEAD

You may need to change the "origin" part to match whatever remove you have set for https://github.com/qmk/qmk_firmware.git, but aside from that .... the above command s should let you test out the code change in question.

The alternative, is to use something like hub, which is a wrapper for git for interfacing with github.

[St0RM53]

@drashna @zvecr
Updated keyboard to work on latest QMK 0.8.133
Test results after enabling OLED driver in rules.mk :

Before applying the fix: https://pastebin.com/GcddgNbj
After applying the fix: https://pastebin.com/SA8ve5X0

Less errors but still an issue.
Here is the keyboard since i have issues updating by branch properly in github: https://www.dropbox.com/s/wcw5xuxjabdpdoz/lilith101.zip?dl=0

Remember to enable oled driver for testing.

[St0RM53]

After help from @zvecr compilation is successful for both I2CD1 and I2CD2; however only I2CD1 was functional in my testing.

[St0RM53]

update: All working fine for I2CD2 as well, the issue was missing definitions for the GPIO bank used for I2C1 pins as i2c_master.h uses by default Bank B where I2C0 pins are set.

[laughingrice]

Everything is working now. Not sure of the exact problem, for some reason this code was much more sensitive to running through a logic level converter than the teensy Arduino library version.

The scope showed that it's running at a much higher default frequency and probably uses a different logic level for 0/1 (logic level shifter was not shifting to the full range for some reason). I removed the logic level converter and am running the passive side (mpc23018) on 3.3 volts (cost a little on the maximum led level) and it's working now.

Looks like my PCB needs to go through another design iteration ...

[St0RM53]

@laughingrice Yes everything is working as expect now. Teensy 3.2 runs on 3.3V so i am assuming your logic shifter might not be designed for that frequency or it's input impedance is too low for the T3.2 output

@drashna I'm assuming this will be closed automatically as soon as it is merged?