Steinar H. Gunderson

In at least DDR Extreme AC, chunk 0x05 contains background scripts.

Without having checked on real hardware: I would suppose it is both correct and incorrect. Ie., most likely, there is a separate lighting chunk, but the data is always identical...

I fixed QC35 (and potentially others) in the Bluetooth_Headset_Battery_Level repository, so the submodule needs to be updated (and a new release made).

I noticed today that this also makes CasparCG output incorrect premultiplied alpha. If you have a white (255,255,255) at 50% alpha, it should be (188,188,188) after premultiplication, not (128,128,128), since...

I also hit the statx() one… that's pretty confusing. Seemingly no way to do fstat() on an opened direct descriptor, then; even disregarding correctness, you'll have to choose between doing...

The problem is that you can't link an open+statx together then, no? You'd have to go a trip through userspace to get the fd return from open registered and into...

Perhaps something like this: ```diff diff --git a/src/apps/fathom.c b/src/apps/fathom.c index e58406b..2e610ea 100644 --- a/src/apps/fathom.c +++ b/src/apps/fathom.c @@ -326,9 +326,9 @@ static void move_to_str(const struct pos *pos, unsigned move, char *str)...

I think it's just super-picky on specific USB IDs, unfortunately. I do have a E3272 (bought an unlocked one off of eBay), but it doesn't even manage to mode-switch it...

Hm, do you have a file I should look at for the lists? Or is it just the regular .rules files?

Aha: ```rust // Don't create a MatchedPos for empty positions at the start // or end. We still want empty positions in the middle of // multiline atoms, as a...