#
…

2026-07-26

00:00hdbngr has quit [Ping timeout: 480 seconds]
00:07hdbngr has joined #asahi-dev
00:15hdbngr has quit [Ping timeout: 480 seconds]
00:23hdbngr has joined #asahi-dev
00:31hdbngr has quit [Ping timeout: 480 seconds]
00:33Carlos has quit [Ping timeout: 480 seconds]
00:33Carlos has joined #asahi-dev
00:39madmatt067 has quit [Quit: .]
00:41hdbngr has joined #asahi-dev
00:42madmatt0677 has joined #asahi-dev
00:55hdbngr has quit [Ping timeout: 480 seconds]
01:04hdbngr has joined #asahi-dev
01:12hdbngr has quit [Ping timeout: 480 seconds]
01:19hdbngr has joined #asahi-dev
01:28hdbngr has quit [Ping timeout: 480 seconds]
01:35hdbngr has joined #asahi-dev
01:44hdbngr has quit [Ping timeout: 480 seconds]
01:47lion328 has quit [Ping timeout: 480 seconds]
01:47lion328 has joined #asahi-dev
01:48ipatch has joined #asahi-dev
01:52hdbngr has joined #asahi-dev
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-dev
02:00hdbngr has quit [Ping timeout: 480 seconds]
02:06hdbngr has joined #asahi-dev
02:15hdbngr has quit [Ping timeout: 480 seconds]
02:19chadmed has quit [Quit: Konversation terminated!]
02:22hdbngr has joined #asahi-dev
02:31hdbngr has quit [Ping timeout: 480 seconds]
02:31chadmed has joined #asahi-dev
02:39hdbngr has joined #asahi-dev
02:52hdbngr has quit [Ping timeout: 480 seconds]
03:01hdbngr has joined #asahi-dev
03:05jmax[m]🎉 to dcp on 14.8.3 fw! This is on M2 currently?
03:08lion328 has quit [Ping timeout: 480 seconds]
03:09hdbngr has quit [Ping timeout: 480 seconds]
03:17nora_ has joined #asahi-dev
03:18hdbngr has joined #asahi-dev
03:18nora has quit [Ping timeout: 480 seconds]
03:20chadmedjmax[m]: nah this is m1 pro
03:26hdbngr has quit [Ping timeout: 480 seconds]
03:31chadmed has quit [Quit: Konversation terminated!]
03:35hdbngr has joined #asahi-dev
03:35chadmed has joined #asahi-dev
03:44hdbngr has quit [Ping timeout: 480 seconds]
03:51hdbngr has joined #asahi-dev
04:00hdbngr has quit [Ping timeout: 480 seconds]
04:02hdbngr has joined #asahi-dev
04:07lion328 has joined #asahi-dev
04:11hdbngr has quit [Ping timeout: 480 seconds]
04:19hdbngr has joined #asahi-dev
04:21lion328 has quit [Ping timeout: 480 seconds]
04:28hdbngr has quit [Ping timeout: 480 seconds]
04:37hdbngr has joined #asahi-dev
04:37svennice!!
04:40JamesCalligeros[m]edid is still busted for some reason. The command hasnt moved or changed size but it just doesnt work. the av service might not be started correctly
04:45hdbngr has quit [Ping timeout: 480 seconds]
04:52hdbngr has joined #asahi-dev
04:57svenalright, time to see why EL0 wants to update pagetables with simd stores now
05:01hdbngr has quit [Ping timeout: 480 seconds]
05:08hdbngr has joined #asahi-dev
05:17hdbngr has quit [Ping timeout: 480 seconds]
05:22svenit’s real EL0 and not my fake GL0 which makes sense because ppl was gl1 only back in that xnu release
05:22svenso probably some corruption somewhere
05:24hdbngr has joined #asahi-dev
05:33hdbngr has quit [Ping timeout: 480 seconds]
05:40hdbngr has joined #asahi-dev
05:43svenNot just simd, it’s just the first I notice because the hv doesn’t emulate the store and I accidentally mapped the pages rw for userspace :D
05:49hdbngr has quit [Ping timeout: 480 seconds]
05:57hdbngr has joined #asahi-dev
06:01svenhuh, the stores are entirely weird. so it’s not trying to update pagetables, xnu somehow handed out memory that I still consider pagetables to userspace
06:12hdbngr has quit [Ping timeout: 480 seconds]
06:19hdbngr has joined #asahi-dev
06:28hdbngr has quit [Ping timeout: 480 seconds]
06:35chadmed has quit [Quit: Konversation terminated!]
06:35hdbngr has joined #asahi-dev
06:38chadmed has joined #asahi-dev
06:38chadmedaaaa did the upstream macsmc-power get the thing to change the magsafe led colour
06:42chadmedhuh no i guess the smc just does that automatically now
06:42chadmedthats a shame
06:50hdbngr has quit [Ping timeout: 480 seconds]
06:52hdbngr has joined #asahi-dev
06:58svenit's my dumb cache. i guess this is what i get for being happy that i finally get to use my leetcode skills from years ago to implement a LRU cache for TTBR0 shadow pagetables :D
06:58svenso a process disappears and i just move the pagetables to my cache but still track them. meanwhile XNU frees all that memory and later hands it out to userspace which then does whatever with it
06:58svenand that was the simd store i saw
06:59svenlet's see if i can get a stable boot without that cache and then it's time to trap all TLB maintenance i guess and see once something in ttbr0 is fully invalidate and drop it
07:07yukaout of curiosity what did you add from the booting system yesterday to "EL0 wants to update pagetables with simd stores"?
07:08yukawere you just trying to speed it up?
07:09sventhat one already had that cache, it just ran into the simd crap 9/10 boots
07:10svenright now i rebuild the entire pagetable whenever i see a write anywhere so i figured i should cache the userspace ones because they'll change on every process switch
07:11svenand well... that cache still contained pagetables that XNU long dropped and later handed the memory to userspace for random allocations
07:11sveni think hitting that simd thing was lucky because i accidentally mapped everything RW to userspace and wouldn't have noticed as quickly otherwise
07:12svenit just took ~10 minutes to boot now, let's see how long it survives :D
07:12hdbngr has quit [Ping timeout: 480 seconds]
07:17jannauasahi-7.1.5-2 pushed to github
07:21hdbngr has joined #asahi-dev
07:28svenlooks like it was the cache. this boot has been stable now
07:28sventime to trap tlb maintenance I guess
07:31hdbngr has quit [Ping timeout: 480 seconds]
07:36jvoisin has quit [Ping timeout: 480 seconds]
07:39hdbngr has joined #asahi-dev
07:48hdbngr has quit [Ping timeout: 480 seconds]
07:48yukaDoes anyone have a base M5 at hand to test nvme for me?
07:48yukaM5 Pro has sart v4 and I haven't looked into that yet
07:57hdbngr has joined #asahi-dev
08:14jvoisin has joined #asahi-dev
08:14jannauchadmed: please use inter tag diff mechanism. also it was safe to assume that I'll handle bumping the kernel ebuilds
08:15hdbngr has quit [Ping timeout: 480 seconds]
08:23hdbngr has joined #asahi-dev
08:30svenand with tlb trapping and dropping asids xnu throws away from the cache I get a stable but still slow boot to single core userspace 🥳
08:32hdbngr has quit [Ping timeout: 480 seconds]
08:35chaos_princess has quit [Quit: chaos_princess]
08:35chaos_princess has joined #asahi-dev
08:40hdbngr has joined #asahi-dev
08:49hdbngr has quit [Ping timeout: 480 seconds]
08:56hdbngr has joined #asahi-dev
09:00jvoisin` has joined #asahi-dev
09:03jvoisin has quit [Ping timeout: 480 seconds]
09:05hdbngr has quit [Ping timeout: 480 seconds]
09:06hdbngr has joined #asahi-dev
09:15hdbngr has quit [Ping timeout: 480 seconds]
09:17leiochadmed: EDID being busted feels a bit encouraging for my monitors 240 Hz fix, lol
09:18JamesCalligeros[m]It’s only the avservice function to copy the edid proper that is not working. TimingElements and ColourElements are all correct, which means DCP itself is correctly handling the EDID
09:23hdbngr has joined #asahi-dev
09:28leiosad, some reddit thread suggests 15.2 beta is what fixed 240 Hz on this monitor for macOS
09:29leio(basically 15.x while broken on 14.x still)
09:31hdbngr has quit [Ping timeout: 480 seconds]
09:36JamesCalligeros[m]Maybe once we look at 26/27 itll work :p
09:39leioit works in macOS with 15.x, I haven't gone for the ugly macos yet, but I assume you are referencing the plans to go 14.x instead of 15.x and then skip 15.x to not have too many to support
09:40hdbngr has joined #asahi-dev
09:40yukayou don't care whether the macos is ugly when there is no macos
09:42jannauI think the only device which would benefit from 15.x support is the m3 ultra mac studio. m4* fuos support is busted in all of 15.x
09:45leioI'm saying M1 would benefit from 15.x purely from getting this very specific monitor to run at 240 Hz without trying and failing with EDID hacks due to bug fixes in the fw
09:45leioeven M1*
09:45leiobut people with asahi and this monitor is probably a total of 1 or 2, me included
09:46leioof course same benefit would come from 26.x or 27.x
09:46jannauI don't think M1* based devices would benefit, only M2 Pro/Max/Ultra and later with HDMI 2.1
09:48leioyes, you're right on that aspect
09:48hdbngr has quit [Ping timeout: 480 seconds]
09:50leiobasically benefits from generic bug fixes, this particular one indeed depends on HDMI 2.1 to matter; I also didn't get it into 240 Hz with other possible connectors either back then; but the only so far known bug fix is for EDID handling for this monitor; the other 240 Hz ones of the era were fine on 14.x due to different ways to express it in EDID
09:51Mrmaxmeier17 has quit [Quit: The Lounge - https://thelounge.chat]
09:52Mrmaxmeier17 has joined #asahi-dev
09:57hdbngr has joined #asahi-dev
09:58chadmedtheres also the issue with user-serviced screens having artefacts and other nonsense on pre-15.x dcp firmware
09:58chadmedso yes eventually i would like even M1 to be on something above 15.x
09:58chadmedbut we need to get SPTM and friends sorted before we even think about that
10:06hdbngr has quit [Ping timeout: 480 seconds]
10:07hdbngr has joined #asahi-dev
10:10linuxgemini has quit [Quit: Ping timeout (120 seconds)]
10:10linuxgemini has joined #asahi-dev
10:20yukat6040 has tunable_CIO4PLL_CORE
10:21hdbngr has quit [Ping timeout: 480 seconds]
10:22yukaso I think that will require some changes
10:22chadmed has quit [Quit: Konversation terminated!]
10:22chadmed has joined #asahi-dev
10:23cy8aerjannau: with asahi-7.1.5-2 I still do not have any battery information on my gnome desktop. For gnome it does not have a battery. (mbpro m1pro)
10:24cy8aerI had this with 7.1.5-1 too, 7.0.13-1 works fine.
10:24chadmedCONFIG_MACSMC_POWER=m?
10:25cy8aerchadmed: hm, no. Is it new?
10:25chadmedthe config symbol changed during upstreaming
10:25cy8aerAh, I see. I'll try it...
10:29hdbngr has joined #asahi-dev
10:44nickchanchaos_princess: For your SPMI series how to get spmi bus' interrupt from ADT?
10:45nickchanclaims to be compatible with apple,t8015-spmi which is upstream so I need to take a look
10:45chaos_princess3 of them are listed, the other was guessed
10:46chaos_princessor you mean the controller?
10:46nickchanum, it's maxItems: 1 in the bindings
10:46nickchanthe controller
10:46chaos_princessits listed, the second one
10:46nickchanI only have that upstream so that's fine
10:50chaos_princessnickchan: also, fifo interrupt does not work on m1
10:50nickchanit doesn't compile with clang :(
11:01chaos_princessnickchan: strange, since i was also buidling with LLVM=1
11:01svenwith proper tracking of changes and only updating what actually changed it now takes 1-2 minutes instead of 10+ to reach the login screen 😁
11:01svenstill only on a single core
11:03nickchanclang version 22.1.8 (Fedora 22.1.8-4.fc44) asahi fedora clang even
11:06hdbngr has quit [Ping timeout: 480 seconds]
11:08hdbngr has joined #asahi-dev
11:12cy8aerchadmed: As you may expect: My setup works with CONFIG_MACSMC_POWER=m. Thanks
11:17hdbngr has quit [Ping timeout: 480 seconds]
11:18ddxtanx has quit [Read error: Connection reset by peer]
11:18ddxtanx has joined #asahi-dev
11:25hdbngr has joined #asahi-dev
11:32nickchanchaos_princess: "fifo interrupt does not work on m1" ... I suppose it's like this -> [ 1.383385] spmi spmi-0: failed to wait for RX FIFO not empty
11:33nickchanokay and a11 works in polled mode so same as m1 as far as the patch is concerned
11:33hdbngr has quit [Ping timeout: 480 seconds]
11:42hdbngr has joined #asahi-dev
11:45yukahow bad is polled mode?
11:45yukaCan I use it for tipd (where I don't know the location of all the irq regs) on M5?
11:50jvoisin has joined #asahi-dev
11:50hdbngr has quit [Ping timeout: 480 seconds]
11:53jvoisin` has quit [Ping timeout: 480 seconds]
11:59hdbngr has joined #asahi-dev
12:08chaos_princessno, tipd needs downstream irqs
12:19hdbngr has quit [Ping timeout: 480 seconds]
12:23hdbngr has joined #asahi-dev
12:24chadmedoh we broke charge inhibition
12:29jannaucharge inhibition as in stop charging now or the stop charge threshold stopping charging at a specific charge percentage?
12:30jannauthe second one was dropped during upstreaming and I forgot to re-add it. it is already fixed in the latest tag
12:31chadmedthe latter, plasma seemingly wont do anything without those props
12:32hdbngr has quit [Ping timeout: 480 seconds]
12:32chadmedlol you did what i was about to commit :P
12:32jannaufixed for me in asahi-7.1.5-2. I only noticed because the magsafe led turned green
12:33chadmedit looks like >=mboot-18000 will do that whenever the smc is told to stop charging now too
12:33chadmedunless my battery readings are busted, the magsafe led went green for me at 80% on j314
12:35jannaunot mboot but the system global smc firmware and yes it does that since 26.4 or so
12:36chadmedyes, i just didnt know which system firmware that corresponds to as i have blown away and reinstalled macos on that machine so many times
12:36chadmedwhatever version of m1n1 is on there doesnt know what it is either, only that its whatever shipped with mboot-18000
12:36chadmedanyway that is annoying and i will have to go get some electrical tape
12:40jannaubrightness control or adaptive brightness would be nice but I don't see much of a difference between orange and green
12:40hdbngr has joined #asahi-dev
12:47chadmedit's mostly noticeable in a dark room, the green is perceptually quite a bit brighter than the orange
12:49chadmedits like trying to sleep next to those 2010s clock radios with blue displays
12:55hdbngr has quit [Remote host closed the connection]
12:55hdbngr has joined #asahi-dev
13:06hdbngr has quit [Ping timeout: 480 seconds]
13:10hdbngr has joined #asahi-dev
13:19hdbngr has quit [Ping timeout: 480 seconds]
13:27hdbngr has joined #asahi-dev
13:41jannauofficial arm64 chrome packages are up, at least the .deb one: https://dl.google.com/linux/direct/google-chrome-stable_current_arm64.deb
13:43hdbngr has quit [Ping timeout: 480 seconds]
13:50hdbngr has joined #asahi-dev
13:57chadmed has quit [Remote host closed the connection]
13:57fugi5 has joined #asahi-dev
13:59hdbngr has quit [Ping timeout: 480 seconds]
13:59chadmed has joined #asahi-dev
14:06hdbngr has joined #asahi-dev
14:19svenand with fgt-at-home, aka just patch everything I care about with hvc instead of using TVM/TRVM, I’m down to about a minute now
14:19hdbngr has quit [Ping timeout: 480 seconds]
14:20svenlet’s see about smp now
14:29hdbngr has joined #asahi-dev
14:38hdbngr has quit [Ping timeout: 480 seconds]
14:46hdbngr has joined #asahi-dev
14:47lion328 has joined #asahi-dev
14:55hdbngr has quit [Ping timeout: 480 seconds]
15:03hdbngr has joined #asahi-dev
15:12hdbngr has quit [Ping timeout: 480 seconds]
15:12hdbngr has joined #asahi-dev
15:14yukasome changes to cs-sw to (configurably) auto-enter debugusb and enter dfu
15:17yukaHowever I found that my devices (newer firmwares?) only allow VDM actions if the CS was plugged in at boot
15:18yukaif I do the Ctrl-X (Disconnect) command, it stops working until I manually reboot the DUT
15:21hdbngr has quit [Ping timeout: 480 seconds]
15:28hdbngr has joined #asahi-dev
15:58chaos_princessyuka: are t8132 and t604x using the same phys?
16:03yukachaos_princess I initially assumed so but doesn't seem like it
16:03yukabecause j773s has a new tunable_CIO4PLL_CORE
16:03yukaand no tunable_CIO3*
16:04yukaits compatible is atc-phy,t6040, but not that it means much... t8132 also had a new adt compatible but was basically the same
16:04chaos_princessbut the usb parts are same?
16:04chaos_princesscio is tb/usb4
16:04yukalet me try, although I don't have a t604x with a usb device plugged in
16:05yukaI can at least check if all the drivers probe
16:05yukaif I skip the CIO parts
16:05chaos_princessnot super helpful unfortunately
16:05chaos_princessatcphy does not start to do things unless you plug sth in
16:05yukaYes, I think flokli can plug in a usb-ethernet or something tomorrow
16:06chaos_princessare you remoting into it?
16:06chaos_princessmakes sense then
16:07yukaor maybe if enverbalalic shows up again they can test it too
16:07yukathey had an m4 max macbook
16:16yukat6040 has _way_ less individual tunables
16:16yuka53 on t8132 vs 25 on t6040
16:17chaos_princesswould be annoying if they re-done it in a significant way
16:17flokliyuka: day after tomorrow, Tuesday
16:18yukaack, thank you
16:25yukathere is only two tunables per LN
16:25yukatunable_LN<X>_RX_CFG_TX_OF_RXCLK
16:26yukatunable_LN<X>_RX_EQ_CIO_DFLT
16:26yukaAh sorry that's the only _new_ tunables
16:27yukaNo, that is literally the only LN<X> tunables in there
16:27yukanone of the old ones are in there
16:28yukalet me just post the list
16:28chaos_princessso i guess the first for usb, the second for tb?
16:28yukalol the paste service banned my ip
16:28yukabecause I pasted too much
16:29yuka(t8132)
16:29yuka(t6040)
16:30yukaThis one remains: LN<X>_RX_TOP_CIO_DFLT
16:30yukathe two I posted above are new
16:30yukaand all others are gone
16:31chaos_princesswhat are tunable_DRD_USB31_DBG_DEFAULT and similar?
16:31yukayou mean the value?
16:32chaos_princessno, what do they do, they are not there on m3 gen
16:32yukaNo idea
16:32chaos_princessless "what they do", more "do you write those"
16:33yukaI can't trace this if this is what you mean
16:33yukaSo I don't know if they should be applied or where
16:33chaos_princessok, sorry, i am being confusing
16:33yukaOh wait
16:33chaos_princessdo you apply those on t8132
16:33yukawere they also on t8132
16:33chaos_princessand if no - do things still work
16:33yukaAh, I didn't apply any new tunables
16:33yukaon t8132
16:34yukaand it still worked
16:34yukaI only used the existing t8122 tunables with the alias for ATC0AXI2AF -> ATCAXI2AF
16:34yukaand that worked just fine
16:34yukagood catch that these are new, I didn't realize
16:35yukaAhh wait
16:35yukaI did an oops and grepped too many tunables out of my adt
16:35yukathese are present on M3 Max, but they're not atcphy
16:35yukaand then there are a lot less tunables
16:35yuka(even less)
16:36hdbngr has quit [Ping timeout: 480 seconds]
16:40hdbngr has joined #asahi-dev
16:41chaos_princessoh, they grew tb5 in t6040
16:42chaos_princessso, phy is probably different, but usb part may be same
16:42yuka'grew'=added?
16:44yukaI think it would be useful to have a 'reset-only' mode in atcphy
16:47jackhumphries has joined #asahi-dev
16:48MikeM64 has joined #asahi-dev
16:48yukaI doubt USB3 will work
16:48yukaas-is this will only result in apple,tunable-axi2af, apple,tunable-common-b & apple,tunable-laneX-cio
16:48yukanone of the tunables that would go into apple,tunable-laneX-usb are present
16:49yukaif that means we just don't need them? Idk
16:49hdbngr has quit [Ping timeout: 480 seconds]
16:49yukacan try, I guess
16:49chaos_princesstunable_LN0_RX_CFG_TX_OF_RXCLK ?
16:49yukaah yes
16:49yukalet me stare at the values for a bit
16:56hdbngr has joined #asahi-dev
17:05hdbngr has quit [Ping timeout: 480 seconds]
17:06yukaI mean I can try to do the old trick of reading the whole area from m1n1, and then reading the whole area again from a kext from a booted kext (with/without $device plugged in), and then diffing
17:06yukaBut not today
17:06yuka* from a booted macOS
17:07yuka(fwiw t6050 has the same tunables etc.)
17:07chaos_princesswhat about patching an adt and causing it to instead write a recognizable pattern
17:07yukathat would work too
17:08chaos_princessand tbh, you don't need to read the entire area
17:08chaos_princesslook at the memory ranges in the adt, when it applies tunables, it uses one of them as base
17:08chaos_princessso, make the tunable write a recognizable flag at offset zero, see which area changed
17:08yukamakes sense
17:09hdbngr has joined #asahi-dev
17:09yukathat still requires reading the memory from macos
17:09yukais there some easier way to do that than a kext?
17:09chaos_princesswrite MacRingZero.kext :P
17:09svenwe could really use that hypervisor 🙃
17:10svendon’t think you can read memory from userspace anymore
17:10yukaI mean, I had an m1n1 uartproxy running an a kext, it was just really buggy
17:10yukafor many reasons...
17:10svenbut yeah, I’d write a kext that lets an ioclient or however they are called map and read anything from userspace
17:11yukahmm, how would it interface with the kext?
17:11yukaor would the kext just enable some existing mechanism?
17:12yukaI'm too much of an xnu/mach noob to know how to obtain this port
17:12yukaI tried writing a character device for sending proxy commands from userspace, but it turned out to be very badly documented
17:13yukabut I guess I can read up on this, just not a today thing...
17:14ddxtanx_ has joined #asahi-dev
17:14ddxtanx has quit [Remote host closed the connection]
17:21chaos_princessso, since you are loading a custom kernel... just patch out the check for ==0 in task_for_pid ?
17:22hdbngr has quit [Ping timeout: 480 seconds]
17:25tq1l has joined #asahi-dev
17:26yukaI have honestly never booted a custom kernel without the help of m1n1, so I'll have to look into that as well, but looks like a doable path
17:26yukathe kexts I can load from the running system once sip is disabled
17:26yuka(or, it needs a reboot, but there is some guidance for that)
17:27tq1l has quit []
17:28hdbngr has joined #asahi-dev
17:33pb17 has quit [Ping timeout: 480 seconds]
17:34NickChan[m]chaos_princess: that's one check but you would also need to patch the check for kernel_pmap and probably the check when converting the task port
17:34NickChan[m]to do stuff like that on macOS I think people uses a kernel read write kext instead of patching that
17:37hdbngr has quit [Ping timeout: 480 seconds]
17:40yukadid anyone already find out what was the last KDK to contain debug symbols?
17:41hdbngr has joined #asahi-dev
17:43yukaor maybe I misunderstood and the .dSYM directory is unnecessary for the development kernels
17:45yukaseems like it, at least I have symbol names
17:47pb17 has joined #asahi-dev
17:48Lyude has quit [Quit: Bouncer restarting]
17:50Lyude has joined #asahi-dev
17:58ddxtanx_ has quit [Quit: Konversation terminated!]
17:59ddxtanx_ has joined #asahi-dev
17:59hdbngr has quit [Ping timeout: 480 seconds]
18:08hdbngr has joined #asahi-dev
18:13ddxtanx_ has quit [Ping timeout: 480 seconds]
18:14jackhumphries has quit [Read error: Connection reset by peer]
18:22hdbngr has quit [Ping timeout: 480 seconds]
18:31hdbngr has joined #asahi-dev
18:42hdbngr has quit [Remote host closed the connection]
18:43hdbngr has joined #asahi-dev
18:52ddxtanx_ has joined #asahi-dev
18:58hdbngr has quit [Ping timeout: 480 seconds]
18:59jackhumphries has joined #asahi-dev
19:04hardy94 has joined #asahi-dev
19:06hdbngr has joined #asahi-dev
19:08ddxtanx has joined #asahi-dev
19:08ddxtanx_ has quit [Read error: Connection reset by peer]
19:16ddxtanx has quit [Ping timeout: 480 seconds]
19:16Carlos has quit [Ping timeout: 480 seconds]
19:17Carlos has joined #asahi-dev
19:17ddxtanx_ has joined #asahi-dev
19:19ddxtanx_ has quit [Read error: Connection reset by peer]
19:19ddxtanx has joined #asahi-dev
19:25hdbngr has quit [Ping timeout: 480 seconds]
19:31ddxtanx has quit [Read error: Connection reset by peer]
19:35hdbngr has joined #asahi-dev
19:35lion328 has quit [Ping timeout: 480 seconds]
19:36ddxtanx_ has joined #asahi-dev
19:43hdbngr has quit [Ping timeout: 480 seconds]
19:48lion328 has joined #asahi-dev
19:54hdbngr has joined #asahi-dev
19:59ddxtanx_ has quit [Ping timeout: 480 seconds]
20:02hdbngr has quit [Ping timeout: 480 seconds]
20:03ddxtanx_ has joined #asahi-dev
20:10hdbngr has joined #asahi-dev
20:21ddxtanx has joined #asahi-dev
20:21ddxtanx_ has quit [Read error: Connection reset by peer]
20:24ddxtanx_ has joined #asahi-dev
20:24ddxtanx has quit [Read error: Connection reset by peer]
20:24hdbngr has quit [Ping timeout: 480 seconds]
20:28ddxtanx has joined #asahi-dev
20:32ddxtanx_ has quit [Ping timeout: 480 seconds]
20:33ddxtanx has quit [Read error: Connection reset by peer]
20:34hdbngr has joined #asahi-dev
20:40ddxtanx_ has joined #asahi-dev
20:43hdbngr has quit [Ping timeout: 480 seconds]
20:45seb_ has joined #asahi-dev
20:46seb_ has quit []
20:48hdbngr has joined #asahi-dev
20:50ddxtanx has joined #asahi-dev
20:51ddxtanx_ has quit [Ping timeout: 480 seconds]
20:53chadmed has quit [Remote host closed the connection]
20:54chadmed has joined #asahi-dev
21:02enverbalalic has joined #asahi-dev
21:04enverbalalicyuka: you got a branch for the USB tests you need me to do ?
21:04enverbalalicm4-usb or t604x-usb-broken ? :D
21:06hdbngr has quit [Ping timeout: 480 seconds]
21:06svenhah, with a bigger shadow pagetable cache this is actually almost usable all the way to userspace even with just a single core running
21:06ddxtanx has quit [Read error: Connection reset by peer]
21:06ddxtanx has joined #asahi-dev
21:06svenand i didn't even need the fancy shadow tlb thingy, just repeating every PTE write to the guarded/non-guarded shadow seems to be fast enough
21:11ddxtanx has quit [Quit: Konversation terminated!]
21:11ddxtanx has joined #asahi-dev
21:14hdbngr has joined #asahi-dev
21:14yukaenverbalalic yeah realized it was more different than expected, so don't bother for now. But I have nvme for you (use at your own risk!)
21:15yukaI'm actually installing a stage2 m1n1 on my mac mini right now
21:20ddxtanx has quit [Ping timeout: 480 seconds]
21:23enverbalalicuu that's interesting
21:24enverbalalici also have to fix the watchdog in linux forcing a reboot every 5min
21:25enverbalalici saw you fixed that on your end yesterday
21:26hdbngr has quit [Ping timeout: 480 seconds]
21:28yukaI fixed some reboot for t8132
21:29yukabut if you know which always-on is missing for that on t6040 I will take that
21:30ddxtanx has joined #asahi-dev
21:33enverbalalic has quit [Ping timeout: 480 seconds]
21:34hdbngr has joined #asahi-dev
21:36ddxtanx_ has joined #asahi-dev
21:39pb17 has quit [Ping timeout: 480 seconds]
21:39pb17 has joined #asahi-dev
21:39ddxtanx has quit [Ping timeout: 480 seconds]
21:40ddxtanx has joined #asahi-dev
21:44ddxtanx_ has quit [Ping timeout: 480 seconds]
21:49ddxtanx has quit [Ping timeout: 480 seconds]
21:51hdbngr has quit [Ping timeout: 480 seconds]
21:51svenlol, as expected safari just dies immediately because i ignore the JIT SPRR crap entirely
21:51svenunfortunately SMP also dies very soon after xnu brings up the 2nd core :(
21:52ddxtanx has joined #asahi-dev
21:56enverbalalic has joined #asahi-dev
21:57Mrmaxmeier17 has quit [Quit: The Lounge - https://thelounge.chat]
21:57ddxtanx has quit [Read error: Connection reset by peer]
21:59hdbngr has joined #asahi-dev
22:00ddxtanx has joined #asahi-dev
22:04hardy94 has quit [Remote host closed the connection]
22:04ddxtanx_ has joined #asahi-dev
22:04ddxtanx has quit [Read error: Connection reset by peer]
22:05ddxtanx_ has quit []
22:05ddxtanx_ has joined #asahi-dev
22:13yukaI'm confused: a m1n1.bin can be installed and booted just fine. then, I add some parameters in the end (chainload etc), and kmutil it again, and then iboot reliably panics
22:13yukais there some special trick for the stage1 parameters?
22:13yukathe same m1n1.bin chainloads just fine too
22:14hdbngr has quit [Ping timeout: 480 seconds]
22:18ddxtanx_ has quit [Read error: Connection reset by peer]
22:18ddxtanx_ has joined #asahi-dev
22:18yukathe installer adds trailing zero bytes
22:18hdbngr has joined #asahi-dev
22:19enverbalalici just started getting random iboot panics when rebooting from linux via macvdmtool
22:19enverbalalici updated to the latest golden gate beta last night
22:19yukaAh, I get those too on t8132
22:19yukajust on normal m1n1 boot
22:19yukawhich is why I use 26.6RC
22:22enverbalalicm4-m5-nvme is a panic on my end
22:22enverbalalici picked up the 3 nvme-apple commits and the t604x nodes
22:23enverbalalici need to pick up some m1n1 stuff as well ?
22:24ddxtanx_ has quit [Read error: Connection reset by peer]
22:27hdbngr has quit [Ping timeout: 480 seconds]
22:31enverbalaliclooks like another locked power domain maybe
22:32yukaenverbalalic what is L2C_ERR_ADR?
22:33yukamore likely I fucked up the device tree
22:33yukawait no, it worked on the t6040 remotely
22:36hdbngr has joined #asahi-dev
22:36yukaenverbalalic you can independently try this m1n1 branch: https://github.com/yuyuyureka/m1n1/commits/feature/t8132-nvme/ and then see if nvme_init() and nvme_read() work
22:37yukabut normally the m1n1 change shouldn't impact the linux change (m1n1 doesn't bring up the nvme unless it's told to)
22:38yukamake sure you get the one with "fix cdw12 value to read 1 block", I just pushed it
22:41enverbalalicL2C_ERR_ADR = 0x300600502288000
22:45hdbngr has quit [Ping timeout: 480 seconds]
22:51enverbalalicwell, nvme in m1n1 is not complaining
22:53hdbngr has joined #asahi-dev
23:02hdbngr has quit [Ping timeout: 480 seconds]
23:09hdbngr has joined #asahi-dev
23:11aites[m]Heya! chaos_princess chadmed thanks for the comments, I tried to address your comments and did a minor refactor, do take a look at the new commit. If the overall logic and structure looks good to you, I'll fix all the syntax style and checkpatch comments proper commit messages and things in it as welll .. I wouldn't want to comb through the checkpatch if I was gonna delete all that code anyways in the end 😅
23:13chaos_princessthe kmemdup thing, is it for dma-safety?
23:13chaos_princessor what is that for
23:14chaos_princessand does the actuator expose a sane hid descriptor
23:15Ardu0us has quit [Quit: %Reality broken: (I)gnore, (R)etry, (A)bort%]
23:16Ardu0us has joined #asahi-dev
23:23hdbngr has quit [Ping timeout: 480 seconds]
23:31hdbngr has joined #asahi-dev
23:43hdbngr has quit [Ping timeout: 480 seconds]
23:49ddxtanx_ has joined #asahi-dev
23:52hdbngr has joined #asahi-dev
23:52enverbalalicyuka: serror was due to different labeling of pmgr in my and your dts, pmgr and pmgr1 switched places. fixing that on my end makes it not serror anymore
23:52enverbalalicbut no /dev/nvme* still
23:54ddxtanx_ has quit [Read error: Connection reset by peer]
…