Harald

I now have a version that works with gcc 6.3 ...but not ready yet...see at the end... Based on my guess, that there must be some quirk in the initialization...

I first thought 25Mhz could be wrong. I think this depends on init. But the other init also creates a 25Mhz internal SSP clock. What I mean is, if I...

here some data from the current calculation (CRf is the real factor, which is (CR0>>8)&0xFF+1 so that real_f = sspclk/CPSR/CRf): ``` old clock: spi: f=25000 sspclk=25000000 delay=1000 CPSR=232 CRf=4 ->...

Ok...just had a look...that was easy. I had added a short delay to setleds(), which isn't good if each sector is setting the leds (even if they don't change). Because...

you may look into my repo, on branch hg42 for a lot of other code changes... if you are interested, I can create some more PRs for each feature (after...

> but this doesn't: > sspr->CR0 = ... given that the other function also uses `SSPx->CR0 = tmp`, the assignement cannot be the problem in itself. So even if it...

Today, I had the same effect in the new functions. Not sure what's going on. At least the `&=~ |=` sequence solved this, too. While testing this, I noticed when...

FYI: to make this work for (probably) all lua you could add: ``` -- make unpack independent of version local unpack = table.unpack or unpack ```

I think, I'll hijack this issue, because I have the same issue. I think it started some weeks ago (I cannot say exactly, I had no spare time, so I...

Ok, I tried some older versions of `python3-dbus.mainloop.pyqt5` (restricting myself to xenial and Debian, because that usually works with kxstudio), but they all need python 3.5 and I have python...