OLED Driver Feature

[XScorpion2]

Standalone OLED Driver based on PR #5151 but without the conversion of every keyboard. This is an opt in feature for keyboard owners to adapt to at their leisure, with the Sol being converted as an example (plus it's what I use myself, so easy to test daily)

Description

This is a standalone OLED Driver for ssd1396 based OLED Displays. The goals of this driver were to provide support on both AVR and ARM boards with faster performance and smaller size than what existed today as well as fix issues that exist in the current ssd1306 driver and its i2c usage causing lost data & deadlocks. In addition, this driver supports features the current options do not such as scrolling and 90 degree rotation.

By the numbers:
make sol/rev1:brianweyer

type	ssd1306	oled driver	difference
firmware size	26828 bytes	25050 bytes	smaller by 1778 bytes
render time	15 ms	3 ms	faster by 12 ms

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

• Make SSD1036 independent from i2c implementation #4968

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).

[and3rson]

This would be absolutely awesome.

[XScorpion2]

Force push rebase on latest master

[drashna]

It looks like #4994 may compete with this, actually. And that PR may be a better implementation.

@XScorpion2 if you wanted to take a look at that PR? Because, personally, I'd like to see one unified implementation.

[XScorpion2]

@drashna I've looked at that pr a few times before.
It updates the internals of drivers/ssd1306 to use i2c_master instead of i2c.
This fixes the send data error that would occur with the old i2c, and gets the driver using the common i2c_master.

However, it still requires each keyboard to switch their implementation over as most keyboards pulled in ssd1306 and i2c files into their own directory and made changes to them. Which is the same catch with this PR. Though converting existing keyboards to that driver would be faster as it behaves identically to today.

Comparing API and features of the original ssd1306 driver vs this one, this driver requires less code for a user (or keyboard) to write for integration, it's faster, smaller firmware footprint, and has additional features: 90/270 degree rotation support, and scrolling.

[drashna]

Mostly my comment was a crosspost, and to make sure both of you have had eyes on the other PR.

[drashna]

So I decided to test this out ....

~~The logo is stuck on the SOL one. I'm using the CRKBD logo file, and the font file... but .... ~~

Edit: Fixed. Was using the wrong file.

[Lenbok]

Working fine for me on crkbd, corne logo showing happily, my changes (cribbed from @r2d2rogers) are in Lenbok@a118a6d

[drashna]

Well, to be honest, as much as I like optimizing the firmware, I think this is one of those points where functionality is more important.

Would it be possible to have defines to set the rotation, but also have it available as a function/call/check/whatever-the-right-term-is? If so, I think that may be the "best" approach.

[drashna]

There are conflicts now.

[XScorpion2]

@drashna conflicts? where? =P

[XScorpion2]

Well, to be honest, as much as I like optimizing the firmware, I think this is one of those points where functionality is more important.

Would it be possible to have defines to set the rotation, but also have it available as a function/call/check/whatever-the-right-term-is? If so, I think that may be the "best" approach.

@drashna Firmware size increases by ~230 bytes (same with lto on or off) with the define enabled. Quick test converting that to a runtime switch (init param) would be about ~330 bytes. None of the necessary additional rotation checks are in any hot loops so perf hit would be minimal.

Edit: Final results are in, 354 bytes larger for moving 90 degree rotation support to a runtime switch.

[XScorpion2]

Rebased & Compacted commits onto latest master

[drashna]

Also, passes Drashna CI. Been using it for a while. It works very well.

[skullydazed]

Can you remove the user keymap changes?

[XScorpion2]

No, because they wanted me update it. I'll get them in here to approve.

[XScorpion2]

[Legonut]

It's got my blessing, good work @XScorpion2

[noroadsleft]

🍌 🚪

😆

[brianweyer]

My keymap looks good. Thanks for the hard work on this one @XScorpion2!

[XScorpion2]

So that's approval from keyboard developer, keymap owners, and removal of potential copyright issues. Anything else or is this good to go?

[skullydazed]

Build failures were unrelated.

Thanks!