#
#asahi-dev · 2026-08-18 15:05
svenit's dumb type-c VDO -> apple bits mapping
#asahi-dev · 2026-08-18 15:05
fl0_idthx, sure
#asahi-dev · 2026-08-18 15:04
svenit'll just be in the code i push later
#asahi-dev · 2026-08-18 15:04
fl0_idsven can you share what was the elusive vse magic?
#asahi-dev · 2026-08-18 14:58
svenstill hates thunderbolt
#asahi-dev · 2026-08-18 14:58
svenalso found out in the process that https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/include/linux/usb/typec_tbt.h#n30 has tbt3 vs. tbt2/legacy swapped
#asahi-dev · 2026-08-18 14:56
svengonna implement that on the linux side and then push it later today
#asahi-dev · 2026-08-18 14:56
svenokay, i think i got the cable vse nonsense down now
#asahi-dev · 2026-08-17 20:48
chadmedoliver: thanks! i figured out that t8112 and t602x can only scan out two interchange planes at a time, if all three planes are in use then one needs to be linear
#asahi-dev · 2026-08-17 19:56
svenhttps://lore.kernel.org/asahi/20260817-b4-tbt-fixes-v1-0-eded2461f5fc@kernel.org/ hopefully that's all the fixes in the core i need
#asahi-dev · 2026-08-17 19:33
oliverchadmed: I am running on interchange with mutter for month now with no dcp crashes. But that's only two planes in use, and one of those (the one for the cursor) is linear iirc. Not sure if that helps you in any way though
#asahi-dev · 2026-08-17 17:00
svenOkay… I’m just gonna go ahead and reverse engineer that cable vse now. can’t really make sense of it otherwise *sigh*
#asahi-dev · 2026-08-17 12:30
jannaumany parts of the driver hold references to the drm device when they ideally shouldn't
#asahi-dev · 2026-08-17 12:28
jannauupstream gpuvm bindings and explicit device lifetime handling vs. years of technical debt
#asahi-dev · 2026-08-17 12:26
svenwhat made the gpu one so painful? lots of rust bindings upstreamed?
#asahi-dev · 2026-08-17 12:26
svenand here i thought thunderbolt was bad already
#asahi-dev · 2026-08-17 12:22
jannaurebasing the gpu driver onto v7.2 is pain. And I pulled in the merge request for 7.3 as well. I suspect it will overall be less pain than doing v7.2 and v7.3 separately
#asahi-dev · 2026-08-17 11:58
tryhard[m]i'm just tryna figure out why rtkit fails to wake up in my pmp v1 driver
#asahi-dev · 2026-08-17 11:51
tryhard[m]can some1 remind me how to track which VA macos maps to in m1n1?
#asahi-dev · 2026-08-17 09:57
yuka(oof)
#asahi-dev · 2026-08-17 09:57
JamesCalligeros[m]american schoolchildren have M16s
#asahi-dev · 2026-08-17 09:56
yukaOh wow I didn't know the iPhone 8 already had an M11. So advanced!
#asahi-dev · 2026-08-17 09:55
NickChan[m]m7 motion coprocessor !!! /j
#asahi-dev · 2026-08-17 09:55
svendon't jinx it! :P
#asahi-dev · 2026-08-17 09:54
yukanow there's a new version which will hopefully last for the next 5 years until M10 or so ;)
#asahi-dev · 2026-08-17 09:54
yukaIt stayed the same for ~5 years, I think qualifies as mostly-stable
#asahi-dev · 2026-08-17 09:54
NickChan[m]though I suppose it's cursed in reality given how the ways it has been broken in the past
#asahi-dev · 2026-08-17 09:53
NickChan[m]It ought to be mostly stable otherwise the iBoot1->iBoot2 handoff for the boot logo would quickly become very cursed
#asahi-dev · 2026-08-17 09:53
sven🙃
#asahi-dev · 2026-08-17 09:52
JamesCalligeros[m]Perchance
#asahi-dev · 2026-08-17 09:52
LarstiQfor the blog post?
#asahi-dev · 2026-08-17 09:51
JamesCalligeros[m]Oh come on i just finished writing about how mostly-stable this stuff is :p
#asahi-dev · 2026-08-17 09:41
yukahttps://termbin.com/rwlpk < rb->buf
#asahi-dev · 2026-08-17 09:41
svenpeter[m]maybe just afk_26.c if the changes are bad enough? 😕
#asahi-dev · 2026-08-17 09:40
yukaafk_rb_hdr is now double the size and epic_sub_hdr does not exist in this form anymore
#asahi-dev · 2026-08-17 09:38
chaos_princesswell, thats not great
#asahi-dev · 2026-08-17 09:38
yukaYes, this is M4 / 26.6
#asahi-dev · 2026-08-17 09:38
chaos_princessaop also uses the afk/rbep/epic, and there were no changes in that part of the protocol on m3/14.8.3
#asahi-dev · 2026-08-17 09:37
yukahttps://github.com/AsahiLinux/m1n1/blob/main/src/afk.c
#asahi-dev · 2026-08-17 09:37
chaos_princessafk?
#asahi-dev · 2026-08-17 09:36
yukaI think I can get somewhere with this. afk.c needs significant changes though
#asahi-dev · 2026-08-17 08:39
yukaIf I get an M2 Mac mini, I'll have one of each generation M1-M5
#asahi-dev · 2026-08-17 08:37
JamesCalligeros[m]My only two desktops are production machines otherwise i wouldve caught this :(
#asahi-dev · 2026-08-17 08:31
yukaWow the older mac minis are still (or again?) expensive given that I got my M4 for 579€
#asahi-dev · 2026-08-17 08:27
yukathe outer / boot object m1n1 should have NO_DISPLAY ofc
#asahi-dev · 2026-08-17 08:27
yukaon an M1/M2 Mac mini
#asahi-dev · 2026-08-17 08:27
yukaCan someone trace m1n1-in-m1n1 doing the dcp_iboot experiment?
#asahi-dev · 2026-08-17 08:21
yukaand this is the output of m1n1 trying to initialize it: https://termbin.com/ldxn
#asahi-dev · 2026-08-17 08:08
yukaasc_dram_mask is not present in the adt, so it falls back to dva_offset = 0. is this maybe a problem?
#asahi-dev · 2026-08-17 08:07
yukahttps://termbin.com/2z6kp <-- experiments/dcp_iboot output on M4 mac mini

plus ›