07:59 <chadmed> jannau: did you make changes in DCP to support night colour or is kwin doing something funky?

08:12 <chadmed> oh yep i see in 6e8dbe0

08:13 <chadmed> where do the CTMs come from? i dont think 2500K is meant to make my screen that same shade of yellow-green the simpsons were for a bit in the mid 90s...

08:13 <chadmed> so something is off

08:15 <chadmed> everything goes this lurid shade of 70s glassware/simpsons yellow-green, and changes dont apply until the brightness is changed

08:21 <chadmed> oh i can see why the latter happens

08:47 <mps> chadmed: so color temperature setting works iiuc?

08:50 <chadmed> no evidently it doesnt :P

08:50 <mps> ah ok

10:02 <Ry_Darcy> Hello all, running my trusty M1 Mini using Glanzmann's 6.3 kernel. I have just noticed that USB4/Thunderbolt is not working anymore. I have an external enclosure with a Samsung 990 Pro I use for backups. This setup worked with the 6.2 kernel. lsusb reports nothing

10:05 <Ry_Darcy> when I plug it into the USB-C port at the back of the M1, nor is their anything in dmesg/syslog. Anyone any thoughts? All kernel options are enabled, the thunderbolt module is being built. Despite modprobing the module, nothing.

10:06 <Ry_Darcy> With previous kernels, dmesg was reporting "high speed device (10000M)" etc. Strange.

10:08 <chadmed> thunderbolt has never worked also this probably belongs in #asahi-alt

10:19 <Ry_Darcy> Agreed. This post does indeed belong in asahi-alt. Though I would mention same nevertheless.

10:20 <sven> I seriously hope no one shipped a kernel with my half broken thunderbolt support to end end users

10:21 <chadmed> Ry_Darcy: are you absolutely sure it was thunderbolt and not just usb 3.2 gen2 2x2 or whatever the heck its called

10:24 <Ry_Darcy> Yes, this external stuff worked with previous (6.2) kernels, plugged into the right port. lsusb and lsusb -t all reported the right stuff. Now dead. Scratches head..

10:24 <sven> sounds like you’re confused, thunderbolt is not related to lsusb

10:25 <chadmed> usb-c port != thunderbolt, sounds like its just a usb device in which case the usb4 module will do nothing

10:26 <Ry_Darcy> You are correct. But USB4 was active as per dmesg (ring buffer) and syslog and reporting 10000M speed. I can plug this setup into a normal USB port and it is reporting 5000M speed.

10:27 <sven> that still doesn’t make sense. you had a usb3 device from what I can tell and some unsupported kernel broke it 🤷🏻‍♂️

10:27 <Ry_Darcy> Perhaps, I am not sure.

10:27 <sven> I am. please complain to the maintainer of the kernel and not here.

10:28 <chadmed> i dont think anyone tested kwin's CTM implementation, its awful

10:28 <Ry_Darcy> I am not complaining. Subject closed.

10:29 <chadmed> i dont *think* it's our driver doing this, afaict it just applies the CTM as it comes from whatever's provided it

10:51 <Ry_Darcy> My apologies for my confusing observations. It was indeed USB3 and not Thunderbolt. I have reverted to a 6.2 kernel and all is well again.

15:14 <jannau> chadmed: works for me? the dialig/preview is a little bit funky but colors start to look funky below 2.000k

15:20 <jannau> chadmed: have you used kde's night color before on HW with support for gamma? I dont's use it normally. it looked roughly as expected

20:48 <chadmed> jannau: yeah it looks completely different to what i expect to see on other hardware and completely different to how macos behaves on the same screen

20:53 <chadmed> i had a play with iomfb_template.c last night and everything there in iomfb_flush is fine so i think its just the interaction of kwin's CTM calculation and lack of working colour management

20:53 <chadmed> it also behaves oddly in that the matrix is only applied on a full flush which kwin seems to never do of its own volition, only when changing the brightness