#
…

2026-08-31

00:15lion328 has quit [Remote host closed the connection]
00:18lion328 has joined #asahi-dev
00:29bennofs3 has quit [Ping timeout: 480 seconds]
00:33bennofs3 has joined #asahi-dev
00:34voidah has quit [Ping timeout: 480 seconds]
00:42voidah has joined #asahi-dev
01:02user890104_ has joined #asahi-dev
01:10user890104 has quit [Ping timeout: 480 seconds]
01:49croissant` has joined #asahi-dev
01:51croissant` has quit []
01:53croissant` has joined #asahi-dev
01:55croissant` has quit []
01:56croissant has quit [Ping timeout: 480 seconds]
01:59croissant has joined #asahi-dev
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-dev
02:22ralim8 has quit [Quit: The Lounge - https://thelounge.chat]
02:22ralim8 has joined #asahi-dev
02:24tobhe_ has joined #asahi-dev
02:31tobhe has quit [Ping timeout: 480 seconds]
03:04scoopy[m] has joined #asahi-dev
04:00nora has joined #asahi-dev
04:02nora_ has quit [Ping timeout: 480 seconds]
04:07zumi_ has joined #asahi-dev
04:11zumi has quit [Ping timeout: 480 seconds]
04:11zumi_ is now known as zumi
04:30voidah has quit [Ping timeout: 480 seconds]
04:39voidah has joined #asahi-dev
04:40dj has joined #asahi-dev
04:45dj has quit []
04:49voidah has quit [Ping timeout: 480 seconds]
04:50LarstiQThis is why the device tree represents the NHI and DART as children of ACIO with addresses relative to that window." curious what the response will be
04:56voidah has joined #asahi-dev
05:05voidah has quit [Ping timeout: 480 seconds]
05:12eli has joined #asahi-dev
05:12voidah has joined #asahi-dev
05:21voidah has quit [Ping timeout: 480 seconds]
05:27sofus has quit [Ping timeout: 480 seconds]
05:30voidah has joined #asahi-dev
05:35sofus has joined #asahi-dev
05:44eli has quit [Quit: Leaving]
05:56svenI think itll be fine, it’s my best guess how the hardware works
06:41voidah has quit [Ping timeout: 480 seconds]
06:50voidah has joined #asahi-dev
07:04rosalyntg_ has quit [Read error: Connection reset by peer]
07:05rosalyntg has joined #asahi-dev
07:18voidah has quit [Ping timeout: 480 seconds]
07:23voidah has joined #asahi-dev
07:24hardy94 has joined #asahi-dev
07:28chadmed has quit [Quit: Konversation terminated!]
07:28chadmed has joined #asahi-dev
07:30fossdd has quit [Remote host closed the connection]
07:31fossdd has joined #asahi-dev
07:31chadmedsofus: can you please sanity-check my dcp/interchange branch with avd? i think it should be correct now
07:32voidah has quit [Ping timeout: 480 seconds]
07:39voidah has joined #asahi-dev
07:54voidah has quit [Ping timeout: 480 seconds]
08:04voidah has joined #asahi-dev
08:12voidah has quit [Ping timeout: 480 seconds]
08:20voidah has joined #asahi-dev
08:32voidah has quit [Ping timeout: 480 seconds]
08:41voidah has joined #asahi-dev
08:44user890104_ is now known as user890104
08:49voidah has quit [Ping timeout: 480 seconds]
08:58voidah has joined #asahi-dev
09:07voidah has quit [Ping timeout: 480 seconds]
09:16voidah has joined #asahi-dev
09:24voidah has quit [Ping timeout: 480 seconds]
09:25voidah has joined #asahi-dev
09:25sofuschadmed: i have time to test later, the meta_tile_{w,h} is sligtly different, you do DIV_ROUND_UP(roundup_pow_of_two(w), layout.tile_w);
09:25sofuswhile i do roundup_pow_of_two(DIV_ROUND_UP(w, tile_dim));
09:26chadmedhm, AIL initialises compression buffers the other way around
09:26chadmedeh if it doesnt work ill swap it, just sent the format mod patch and cced you anyway
09:33voidah has quit [Ping timeout: 480 seconds]
09:35jmax[m]nit: 's/Unlinke/Unlike/g' in the patch
09:37jmax[m]Probably not worth a v2 :D
09:39JamesCalligeros[m]Lol sashiko caught it too
09:39JamesCalligeros[m]I dont expect a v2 to be required but ofc i will fix it if it is
09:41voidah has joined #asahi-dev
09:46picnoir has quit [Ping timeout: 480 seconds]
09:49voidah has quit [Ping timeout: 480 seconds]
09:54picnoir has joined #asahi-dev
09:58voidah has joined #asahi-dev
10:06zumi_ has joined #asahi-dev
10:06voidah has quit [Ping timeout: 480 seconds]
10:10zumi has quit [Ping timeout: 480 seconds]
10:11zumi_ is now known as zumi
10:15voidah has joined #asahi-dev
10:27voidah has quit [Ping timeout: 480 seconds]
10:35pb17 has quit [Ping timeout: 480 seconds]
10:51pb17 has joined #asahi-dev
11:16voidah has joined #asahi-dev
11:31voidah has quit [Ping timeout: 480 seconds]
11:52voidah has joined #asahi-dev
11:56hardy94 has quit [Quit: hardy94]
12:06bennofs3 has quit [Ping timeout: 480 seconds]
12:11bennofs3 has joined #asahi-dev
12:15yurix has quit [Remote host closed the connection]
12:15voidah has quit [Ping timeout: 480 seconds]
12:16voidah has joined #asahi-dev
12:16yurix has joined #asahi-dev
12:30yurix has quit [Remote host closed the connection]
12:30yurix has joined #asahi-dev
12:35oolong has quit [Read error: Connection reset by peer]
12:35oolong has joined #asahi-dev
12:46hardy94 has joined #asahi-dev
12:46hardy94 has quit []
12:46pb17 has quit [Ping timeout: 480 seconds]
12:51pb17 has joined #asahi-dev
12:51balrog_ has joined #asahi-dev
12:54balrog has quit [Ping timeout: 480 seconds]
13:08ddxtanx has quit [Quit: Konversation terminated!]
13:09ddxtanx has joined #asahi-dev
13:15hardy94 has joined #asahi-dev
13:15hardy94 has quit []
13:45picnoir has quit [Ping timeout: 480 seconds]
13:46alessaba has quit [Remote host closed the connection]
13:53picnoir has joined #asahi-dev
14:25voidah has quit [Ping timeout: 480 seconds]
14:44pb17 has quit [Ping timeout: 480 seconds]
14:48alessaba has joined #asahi-dev
14:48nela has joined #asahi-dev
14:51pb17 has joined #asahi-dev
15:09bennofs3 has quit [Quit: Ping timeout (120 seconds)]
15:10bennofs3 has joined #asahi-dev
15:20alessaba has quit [Remote host closed the connection]
15:25nickchanso after some more investigation into rtkit protocol version 10 it seem like the minimum protocol version in the HELLO message when the maximum version is 10 is not actually the minimum protocol version but rather a minor protocol version
15:25nickchanand the inconsistent app endpoints start "within the same protocol version" is the difference between protocol 10.0 and 10.2
15:26nickchan10.0 seen on iOS 10 has app endpoints start at 5 and 10.2 seen on iOS 11-12 has app endpoints start at 6
15:27nickchan10.0 has a bunch of annoying quirks too even when compared to 10.2
15:29nickchani say annoying quirks but it is the AP_PWR_ACK message only being partially implemented so either linux gets no response after setting AP power state (the other side effect like enable syslog seems to exist though), or it gets a response that says it go into state 0x0 when trying to set AP power state to 0x10
15:32nickchans/11-12/11-13/
15:33ipatch has quit [Quit: WeeChat 4.9.5]
15:34ipatch has joined #asahi-dev
15:57nickchancan't say anything about iOS 9 since my mock kernelcache m1n1 doesn't boot with the iBoot, and iOS 10.3.3 is the lowest version I care about (signed on 5s + last version for 32 bit apps)
16:01yukathe other day I said the MAX_PEND_CMDS_CTRL was being set to 64 in the M4 trace
16:01yukait's not true
16:01yukaIt's set to 63 in the M4 trace, and 64 in the A18 Pro trace
16:02yukaAPPLE_NVMMU_NUM_TCBS is set to 63 on both
16:03Amilie has joined #asahi-dev
16:03yukaSo I guess I will introduce another property on apple_nvme_hw
16:04yukahas_zero_based_max_pend
16:04yukaor something like that
16:04yukaAnd then A18 Pro gets its own compatible where that is set to false (like for previous gens)
16:06Amilieo/ all - been looking into DCP as currently M3 Pro Macs run on the FrameBuffer - is there an inprogres / working version of this or is it a currently incomplete thing with no-one working on it? I know that there's the firmware ABI on the WIP branch for 7.2 but there's no DCP nodes for t6030.dsi or t603x-j514-j516.dsi, am I just not seeing these in a branch or are they just
16:06Amilieunwritten? Thanks!
16:07AmilieThank you!
16:07svenyuka: sounds good
16:07chaos_princessthey are mostly unwritten, i have previously tried throwing together something on a t6031, but it did not work first try, and i went to do other things
16:08AmilieI'll take a small dabble but don't expect much as I've not touched DCP before - I'll study the M2stuff then look into the M3 stuff
16:08svenyuka: might also be a good idea to double check if it's really hw-based and not some fw difference
16:09chaos_princessyou should probably use the t602x as reference specifically
16:09chaos_princessa lot of addresses are the same on t602x and t603x
16:09yukaI guess... the M4 was running 26.6 for sure, the A18 Pro might have been 27, need to check
16:09sven(i.e. it's 64 on 26.6 but 63 on 27.0 or whateveR)
16:09AmilieAlrighty, thank you - I'll use t602x as a starting point then and go from there
16:09sveni hope it isn't because i was so happy we managed to avoid that apple,firmware-abi mess :(
16:16Amilie has quit [Quit: WeeChat 4.10.0]
16:23tryhard[m]jannau: chaos_princess : i opened a new pr for asahi-wip-7.2: https://github.com/AsahiLinux/linux/pull/589
16:24tryhard[m]not sure whats up with the indentation in git diff tho
16:24chaos_princessyou could have retargeted the original one, no?
16:26tryhard[m]there are some things asahi-wip-7.2 fixes regarding lifetime stuff
16:28jannauis it possible to change the target branch of a PR?
16:29tryhard[m]don't think so. tried before but never showed up as an option
16:30chaos_princessyes, click the pencil near the pr name
16:30tryhard[m]uh you're right, probably it's new
16:34tryhard[m]tryhard[m]: btw it's not new lol
16:34hardy94 has joined #asahi-dev
17:31voidah has joined #asahi-dev
17:50yukaThe trace showing MAX_PENDS = 63 on t8132 was not a trace of macOS at all
17:51yukait was a trace of m1n1 in m1n1
17:51yukam1n1 sets NVME_MAX_PEND_CMDS_CTRL to "((NVME_QUEUE_SIZE - 1) << 16) | (NVME_QUEUE_SIZE - 1))"
17:56yukaI think we're just being inconsistent with how we use the max_queue_size
17:56yukaI think either value will work fine, if we write it to all the places, but I'll try that now
17:57svenwould be nice if it's just another bug :D
17:59yuka"assert failed: [7454]:CQ (Host I/O) DB error, status_reg: 0x4, head: 63, valid_status: 0x4, err_info_0: 0x1, err_info_1: 0x20000"
17:59yukais the full error when writing
18:08hardy94 has quit [Remote host closed the connection]
18:15vimproved has quit [Remote host closed the connection]
18:15vimproved has joined #asahi-dev
18:25yukaI/O CQ : 0x1001880c000, size: 65, tail/head: 64/63
18:26yukais the tail 64 maybe the problem?
18:26sventhat looks wrong, yes
18:27svenwait.. with size 65 that should be okay
18:27yukawhy is size 65 though
18:30sofuschadmed: re meta_tile order i think its actually the same, and as expected it still works
18:30sofusi also wanna test the alignment, i think its 128 or 256 for avd
18:50midou has quit [Remote host closed the connection]
18:51midou has joined #asahi-dev
18:55riker77 has quit [Quit: Quitting IRC - gone for good...]
19:00vimproved has quit [Remote host closed the connection]
19:00riker77 has joined #asahi-dev
19:00vimproved has joined #asahi-dev
19:29veloek has quit [Quit: leaving]
19:49yukasorry about the misinfo earlier, macos sets MAX_PENDS to 0x400040 on both t8132 an t8140
19:49yukalowering the max_queue_size to 63 results in the same errors at size: 64, tail/head: 63/62
19:50yukathere are still a lot of differences in the trace which I can try to apply but I'm done for today
20:00nickchan has quit [Quit: ZNC 1.9.1+deb2+b3 - https://znc.in]
20:00nickchan has joined #asahi-dev
20:24hardy94 has joined #asahi-dev
20:35voidah has quit [Ping timeout: 480 seconds]
20:35voidah has joined #asahi-dev
21:11JamesCalligeros[m]sofus: it should be 128b aligned aiui
21:30yurix has quit [Read error: No route to host]
21:30yurix has joined #asahi-dev
21:56ddxtanx has quit [Remote host closed the connection]
22:19linuxgemini has quit [Quit: Ping timeout (120 seconds)]
22:19linuxgemini has joined #asahi-dev
22:32chadmed has quit [Quit: Konversation terminated!]
22:32chadmed has joined #asahi-dev
22:38ralim88 has joined #asahi-dev
22:38msimonelli2 has joined #asahi-dev
22:44ralim8 has quit [Ping timeout: 480 seconds]
22:44ralim88 is now known as ralim8
22:44msimonelli has quit [Ping timeout: 480 seconds]
22:44msimonelli2 is now known as msimonelli
22:46apc has quit [Read error: Connection reset by peer]
23:04sai02 has quit [Quit: bye02]
23:04k70_ has joined #asahi-dev
23:05sai02 has joined #asahi-dev
23:05k70 has quit [Read error: Connection reset by peer]
23:05k70_ is now known as k70
23:09ddxtanx has joined #asahi-dev
23:13sai02 has quit [Quit: bye02]
23:13sai02 has joined #asahi-dev
23:47hardy94 has quit [Remote host closed the connection]
…