#
✕ nick:leio
#asahi-dev · 2026-11-15 18:38
leiojannau: there are DCP changelogs somewhere?
#asahi-dev · 2026-11-15 18:37
leioah, so with that stack you maybe already get gst vaapipostproc to use it? (and it didn't work before?)
#asahi-dev · 2026-11-15 18:29
leiosofus: so that wip driver of yours exposes a v4l2 device which gstreamer will expose as a v4l2${basename}convert element? :)
#asahi-dev · 2026-11-15 18:28
leioI don't recall if I had any asahi problems with it though once the clock limit was removed from kernel driver
#asahi-dev · 2026-11-15 18:26
leionewer DCP*
#asahi-dev · 2026-11-15 18:26
leioI'm on Samsung G80SD now with some newer needed to do 240 Hz
#asahi-dev · 2026-11-15 18:26
leiothat's the one with the flickering backlight that I returned 3 panels of
#asahi-dev · 2026-11-15 18:25
leiothat'd be my wild guess for R meaning then
#asahi-dev · 2026-11-15 18:18
leiosofus: does the MSR thing also do rotation?
#asahi-dev · 2026-10-16 10:04
leio(fourcc being essentially four letters iirc :)
#asahi-dev · 2026-10-16 09:53
leiowouldn't fourcc need to consider that other platforms might have their own different interchange format
#asahi-dev · 2026-10-14 08:40
leiomy legacy stuff returns 200 for everything except 5xx's with actual 4xx code in json payload it's great :D
#asahi-dev · 2026-10-13 09:52
leioare you going to work on the code and refactoring next that makes it actually do the installation? :)
#asahi-dev · 2026-10-12 09:05
leiofwiw, on second thought, it might be possible to nerdsnipe me into some appkit/swiftui thing if we are talking about own notarization anyways
#asahi-dev · 2026-10-12 08:54
leioyeah, as long as the python code is refactored to be shared, so it's just different output modes of some kind or something, yeah
#asahi-dev · 2026-10-12 08:53
leioif the installer is a web server that internally talks to the CLI stuff and exposes HTML that we can just show in safari, that's fine, I'm thinking of the parts for sharing code. Ultimately whoever does the work gets to pick what's best, I assume :)
#asahi-dev · 2026-10-12 08:51
leiosorry, I'm talking about how the CLI stuff remains and shares code; if the installer itself is a web interface, but the new backend part of it is what talks to such a wizard-like shared code with CLI, sure
#asahi-dev · 2026-10-12 08:50
leioI still think a web interface isn't a good fit, as it's more of a wizard alike flow in the CLI too, so you'd just be coding that kind of stuff in there, might as well just make the existing CLI steps programmatically better to read-write. Like how do you expose an API there that makes sense and isn't just shoehorning wizard progress contextual stuff assumptions in it
#asahi-dev · 2026-10-10 14:08
leiochaos_princess: there's fastapi too as a modern contender to flask in these situations apparently, fwiw
#asahi-dev · 2026-10-10 11:24
leio a prompt with what text it is.
#asahi-dev · 2026-10-10 11:24
leiomy immediate thought would be that the CLI app has an option to input and output stuff in a more programmatical manner (e.g. JSON), and then a GUI app would just launch the CLI in that mode and control it via stdin/stdout of it that way, converting the output from it to graphical prompts and the answers also accordingly; basically the CLI in that mode could output its prompts in a semi-standardized (between it and the GUI) JSON payload that just tell what kind of
#asahi-dev · 2026-10-10 11:06
leioaw, I daydreamed of RIIRing it all *g
#asahi-dev · 2026-10-10 11:03
leioI'm a native GUI toolkit person; problem is, that's GTK to me, though its macOS support is good-ish, just pointless to ship all that too. But lets not go daydreaming yet :) If GUI, I would probably either go with GTK regardless, a Swift macos native thing or maybe slint/iced/etc.
#asahi-dev · 2026-10-10 10:59
leioor maybe an non-LLM fancy graphical installer to take away a certain other supposed potential marketing point :P
#asahi-dev · 2026-10-10 10:58
leioor maybe I just have something that I can do then (decryption support)
#asahi-dev · 2026-10-10 10:57
leioyes, I would not put it in my initial macos volume
#asahi-dev · 2026-10-10 10:57
leio(I think I had the installer prep the partitions for me though, but doesn't matter)
#asahi-dev · 2026-10-10 10:56
leiook, no problem, I did that for my M2 Ultra way before it was supported in installer; so in that case upgrade of macOS doesn't matter
#asahi-dev · 2026-10-10 10:55
leioI'm confused now due to "<yuka> Installer is nonfunctional with macOS 26 atm"
#asahi-dev · 2026-10-10 10:54
leiofrom the above I read I shouldn't let the macOS on it upgrade to 26?
#asahi-dev · 2026-10-10 10:54
leiogreat, I'll ask which branches once I have had time to set up the basics on it, now that I know it's worth the time spend by now :)
#asahi-dev · 2026-10-10 10:52
leiodev is catching up with gentoo packaging stuff (including any asahi things, but mostly other things) and looking at AVD stuff over holiday break
#asahi-dev · 2026-10-10 10:51
leioI want to use it for development, but including packaging work, which would just need working storage, all CPU cores and ideally working USB for ethernet dongle and what is needed to get that point (display and keyboard, I suppose).
#asahi-dev · 2026-10-10 10:49
leioI'm used to chainload kernel+dts as m1n1 payload
#asahi-dev · 2026-10-10 10:49
leioHow would a J715 M4 base air fare by now?
#asahi-dev · 2026-10-01 06:43
leioLarstiQ: it went like https://www.youtube.com/live/TPwe53kCJuM?si=YiAjAQa7R2Sm0Ejh&t=2058
#asahi-dev · 2026-09-24 10:26
leioMaybe ndufresne has some thought about the above when he wakes up
#asahi-dev · 2026-09-22 14:34
leiosofus: yeah, ok, sounds good re loa dbalancing
#asahi-dev · 2026-09-22 13:17
leiowas wondering the impact of that in what gets moved into CM3
#asahi-dev · 2026-09-22 12:49
leioon Ultra SoCs with that 2 AVD blocks, is the CM3 shared
#asahi-dev · 2026-09-19 12:24
leiomaybe it's not a very outlandish idea, actually, pair it as a perk for a huge donation valued higher than the resale cost or something
#asahi-dev · 2026-09-17 12:39
leiotracked*
#asahi-dev · 2026-09-17 12:39
leioCargo.lock looks tracker in git, shouldn't that get updated too then?
#asahi-dev · 2026-09-15 10:09
leioHow are userspace things currently faring with encountering a vulkan device that doesn't expose those queues, or that assume the same first iterated device would do VV too
#asahi-dev · 2026-09-15 10:05
leiolinkmauve: hey, your v4l2 VV thing, is it on top of mesa or separate?
#asahi-dev · 2026-09-06 18:07
leio"thunderbold" in cover letter "Changes in v2" (commit 0f3086f40d7), also "insid" -> "inside", don't understand the "Moved port unlock" sentence after that, something is off with it; commit e3cabed33 log probably s/not SVID/no SVID/? In commit bbee15ba6 code comments "The altmode has alreayd" -> already
#asahi-dev · 2026-09-06 18:07
leioWell, now you made me try to find more feedback, but only thing I could really do is nitpicking on English and typos, so feel free to just ignore :)
#asahi-dev · 2026-09-06 17:20
leioit's tb, not tbt! /me hides
#asahi-dev · 2026-08-12 08:24
leiofwiw, the GRO message happens for end0 for 10G Aquantia AQtion AQC113 (Antigua 10G) as well, been idly wondering about that, but no speed issues there at 400 Mbps levels at least
#asahi-dev · 2026-08-08 13:05
leiosofus: github being github and can't load the last commit for me rn, but earlier one probably a s/never/newer/ in comment

plus ›