#
…

2026-08-25

00:04aurora has joined #asahi-dev
00:47test has joined #asahi-dev
00:52test has quit []
00:57test has joined #asahi-dev
00:57testStanfordAppliedCyber[m]: per your SPTM emu (tainted), is any of that public/available? I'm trying to do some re/tracing in macOS on the J700.
01:02test has quit [Quit: bye]
01:08___nick___ has quit []
01:08alessaba has quit [Read error: Connection reset by peer]
01:09alessaba has joined #asahi-dev
01:11___nick___ has joined #asahi-dev
01:11___nick___ has quit []
01:13___nick___ has joined #asahi-dev
01:20vimproved_ has joined #asahi-dev
01:20vimproved has quit [Read error: Connection reset by peer]
01:20vimproved_ is now known as vimproved
01:28chadmed has quit [Read error: Connection reset by peer]
01:29chadmed has joined #asahi-dev
01:29chadmed has quit [Read error: Connection reset by peer]
01:30chadmed has joined #asahi-dev
01:34gargantua has quit []
01:54al3xtjames2 has quit [Quit: al3xtjames2]
01:59ids1024 has joined #asahi-dev
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-dev
02:26maettu102 has quit [Remote host closed the connection]
02:26lederstr- has joined #asahi-dev
02:26maettu102 has joined #asahi-dev
02:29lederstrumpf has quit [Ping timeout: 480 seconds]
02:32tobhe_ has joined #asahi-dev
02:39tobhe has quit [Ping timeout: 480 seconds]
03:08nora has joined #asahi-dev
03:10nora_ has quit [Ping timeout: 480 seconds]
03:13mripard has quit [Remote host closed the connection]
03:15mripard has joined #asahi-dev
03:37Ardu0us has quit [Quit: %Reality broken: (I)gnore, (R)etry, (A)bort%]
04:13Yushi has quit [Remote host closed the connection]
04:30Ardu0us has joined #asahi-dev
04:39NiklasSheth[m] has joined #asahi-dev
04:48ddxtanx has quit [Remote host closed the connection]
04:49ddxtanx has joined #asahi-dev
04:57Yushi has joined #asahi-dev
05:00sefidel has quit [Remote host closed the connection]
05:05sefidel has joined #asahi-dev
05:11Yushi_ has joined #asahi-dev
05:13Yushi has quit [Ping timeout: 480 seconds]
05:15Yushi_ is now known as Yushi
06:18Lyude has quit [Quit: Bouncer restarting]
06:19Lyude has joined #asahi-dev
06:20al3xtjames2 has joined #asahi-dev
06:25lyudess has joined #asahi-dev
06:28Lyude has quit [Ping timeout: 480 seconds]
06:39ddxtanx_ has joined #asahi-dev
06:40ddxtanx has quit [Read error: Connection reset by peer]
06:57ddxtanx_ has quit [Remote host closed the connection]
06:59ddxtanx has joined #asahi-dev
07:01hardy94 has joined #asahi-dev
07:09chadmed has quit [Quit: Konversation terminated!]
07:12maettu1027 has joined #asahi-dev
07:13maettu102 has quit [Read error: Connection reset by peer]
07:13maettu1027 is now known as maettu102
07:14lederstrumpf has joined #asahi-dev
07:16lederstr- has quit [Ping timeout: 480 seconds]
07:24foltz51 has joined #asahi-dev
07:30foltz5 has quit [Ping timeout: 480 seconds]
07:30foltz51 is now known as foltz5
07:30pb17 has quit [Ping timeout: 480 seconds]
07:39pb17 has joined #asahi-dev
08:01chadmed has joined #asahi-dev
08:31jannau_any idea why icmp over usb4net would work but tcp doesn't? I'm testing dart remap on m1 max
08:32jannau_ is now known as jannau
08:32chaos_princesshow long is icmp roundtrip time?
08:33jannauless than 1ms and the connected device should not have any other network interface up
08:35jannautcp connect fails immediately with "Failed to connect to" with the correct ip but on not assigned ip addresses it takes normal timeout of ~3s
08:35chaos_princessfire up 'shark, see what happens there?
08:36chaos_princesswhat are you using to test tcp
08:44jannausolved, it was fedora's firewall
08:45jannauso remap works there as well
08:47jannauwow, I've panicked macos with iperf3 over usb4net
08:50jannauinvalid PTE on dart-acio1 sid 1
09:02jannaureported but not sure if Apple still cares about Sequoia
09:04chaos_princessthey only care about 2 major versions down
09:05svenyuka: do you need any more changes to the m4 device trees? I’ll pick them up after rc1 otherwise
09:06yukaThey are ready
09:06svenok, will pick them up on Sunday then
09:13bennofs3 has quit [Quit: Ping timeout (120 seconds)]
09:13bennofs3 has joined #asahi-dev
09:41al3xtjames2 has quit [Read error: Connection reset by peer]
09:41al3xtjames2 has joined #asahi-dev
10:02bennofs3 has quit [Ping timeout: 480 seconds]
10:07bennofs3 has joined #asahi-dev
10:12bgb has quit [Quit: WeeChat 4.10.0]
10:23picnoir has quit [Ping timeout: 480 seconds]
10:29ralim8 has quit [Quit: The Lounge - https://thelounge.chat]
10:31picnoir has joined #asahi-dev
10:34__nick__ has joined #asahi-dev
10:38___nick___ has quit [Ping timeout: 480 seconds]
10:43nickchanam i missing something, where is apple,m3-mailbox-v2 used in Linux
10:43nickchannot a single mention in the device tree
10:49nickchanthanks am adding support for a mailbox that has some degree irq controls yet distinct from the m3 work model
11:03rzhikharevich has joined #asahi-dev
11:10minigyima has joined #asahi-dev
11:13minigyima has quit []
11:21rzhikharevich has quit [Ping timeout: 480 seconds]
11:29rzhikharevich has joined #asahi-dev
12:01rzhikharevich has quit [Ping timeout: 480 seconds]
12:07ralim8 has joined #asahi-dev
12:10rzhikharevich has joined #asahi-dev
12:18chadmed has quit [Quit: Konversation terminated!]
12:18rzhikharevich has quit [Ping timeout: 480 seconds]
12:22chadmed has joined #asahi-dev
12:34tryhard[m]chaos_princess: where do u think the closure maps_addresses should go to? separate method?
12:34chaos_princessthat is one option
12:35svenhas anyone tested suspend with tbt btw? I wonder if we resume into an SError or if tipd does the teardown correctly
12:38chaos_princesstryhard[m]: just a for-loop over a table of values
12:47jannauI did accidentally at some point with tbt-reset-wip, iirc no wakeup but I'm not sure macsmc-input was enabled
12:47jannauno SError if I'm not mistaken
12:51bennofs3 has quit [Quit: Ping timeout (120 seconds)]
12:51bennofs3 has joined #asahi-dev
12:54jannausven: SError on wakeup in DART (t6020: 0xb01a80080)
12:54svenyeah…. That’s what I’m was afraid of
12:56jannauacio appears to be shutdown on suspend. last rtkit syslog message "yslog message: clx.c:776: CLX CFG: clx_en changed 0x0 --> 0x3" and before that two messages with tear_down
13:04sventhe easy way out is to prevent suspend whenever a tbt connection is active
13:11rzhikharevich has joined #asahi-dev
13:15jannauany complications in calling apple_cio_tbt_switch_set(TYPEC_THUNDERBOLT_SWITCH_OFF) during suspend? either directly (or through tipd, probably not as it shouldn't know that acio needs switched off)
13:19rzhikharevich has quit [Ping timeout: 480 seconds]
13:32jannauonly mac mini and studio refreshed
13:33chadmedi suspect laptops will come on or around the usual september event on a stage
13:33aites[m]DAMN 3 core types ??
13:34jannauno, that's just iphones, laptops are usually october/november
13:34chadmedaites[m]: 3 core types isnt new for m6
13:35chadmednot a problem for us either, other than the marketing names being extremely stupid
13:35jannauit is if I'm not mistaken. m5 had super and efficiency and m5 pro/max super and new performance core
13:36jannauthere are other arm socs with 3 core types though
13:36chadmedohhh i thought the base m5 had ecores still
13:37jannauitr kinda had. when it was introduced the super cores were still called perfomance cores
13:38nickchanthe bigger core in a19 pro is the same type of core as the one in m5, but it is only marketed as "performance core"
13:38chaos_princesschadmed: qcom has a soc with like 4 types
13:39chadmedyo dawg i heard you like core types
13:39chaos_princessprime/perf/last gen perf for aarch32/e
13:39nickchanerm, same type of core as the super core in m5
13:39aites[m]thats cool, I wonder how having 3 core types would affect performance
13:40chadmedit wont really, it will allow the scheduler to make different energy efficiency tradeoffs though
13:40nickchanwe'll need to continue ignoring the marketing in device tree to preserve our sanity with a19 pro
13:40nickchanif there's any sanity left which I think is about none
13:40chadmedif something is juuust too demanding for the ecores the scheduler can put it on the mid cores and avoid waking up the pcores if it doesnt need to
13:41chadmedof course in the real world its not so cut and dry. the pcores at a lower pstate might have better perf/w than the mid cores at a higher pstate
13:41aites[m]how conscious is the current linux scheduler (EEVDF I think?) of these P and E core shennanigeans?
13:41aites[m]or is that more firmware side
13:41jannaum6 mac mini has only from 24GB RAM the full memory bandwidth
13:42chadmedthe scheduler is very cognisant of it
13:42nickchanit should be conscious enough given the amount of androids
13:43chadmedhardware scheduling is more of a gpu thing, though intel had to bring it back for their heterogeneous socs initially because the windows scheduler is so shit
13:43nickchanwe'll be able to confirm whether m5 ultra is still t6050 soon enough
13:43jannaustill tb4 / DP 1.4 on M6
13:43chadmedin the linux world we just tell the scheduler how fast the core is with a meaningless dmips/mhz metric, how much energy it uses at each pstate, and it does the rest
13:43chadmedit can handle an arbitrary number of different cores with different perf
13:44svenjannau: that’s gonna make devices disappear and I’m not sure the pm core will like that…
13:45aites[m]ohh that's chill
13:46jannaudid we notice that M5 Max has 5 DCPs?
13:47chaos_princessm6 also has quite a lot for its size class
13:48jannaunickchan: T6051 or T6052 would be consistent with M1-M3. will have the same MIDR as T6050
13:49jannauerr, the Max variant always had that. DCP and 4 DCPext
13:49nickchanm5 max reports as t6050
13:49nickchanand it does make sense since it's just the gpu die different
13:50jannauyes, m5 pro and max probably share the CPU die, m5 ultra has 2 CPU (and GPU) dies
13:51jannauso it still doubles a Max which resulted in a different chip id in the past
13:52jannauthe mini has a new ethernet nic, 2.5 GBps
13:58nickchanthe pricing
13:58nickchanI guess I should have expected this but it is not cheap
13:58chadmed has quit [Quit: Konversation terminated!]
13:58chadmed has joined #asahi-dev
14:21user890104_ has joined #asahi-dev
14:28user890104 has quit [Ping timeout: 480 seconds]
14:29jannausven: I wonder why it's a problem though. afaik children should suspend before it parent and resume after
14:32user890104_ is now known as user890104
14:32user890104 has quit []
14:32user890104 has joined #asahi-dev
14:34sventhe power domains are turned off
14:34svenand that requires the full bootstrap sequence on resume then
14:37jannauoh, the acio driver has no PM handling at all
14:39lion328 has quit [Ping timeout: 480 seconds]
14:40svenyeah, and I suspect atcphy might just go down due to dwc3 suspend
14:49runxiyu__ has joined #asahi-dev
14:49tryhard[m]chaos_princess: u want to remove APPLE_PMP_V1_REPORT entirely right?
14:50runxiyu_ has quit [Quit: runxiyu_]
14:50chaos_princesseither that or have it hidden
14:52lion328 has joined #asahi-dev
15:00lion328 has quit [Ping timeout: 480 seconds]
15:06lion328 has joined #asahi-dev
15:10kettenisI wonder what the new 2.5 GbE chip on the new mini is
15:20jannaumaybe aquantia like the 10G one? Broadcom has afaik none
15:20jcbm has quit [Remote host closed the connection]
15:21jcbm has joined #asahi-dev
15:51tryhard[m]whats the depth level i need to go into the yaml binding doc?
15:51tryhard[m]in ‎Documentation/devicetree/bindings/
16:10rzhikharevich has joined #asahi-dev
16:28chaos_princessMostly just use the pre-existing upstreamed ones as guidance
16:43rzhikharevich has quit [Ping timeout: 480 seconds]
16:51rzhikharevich has joined #asahi-dev
16:59rzhikharevich has quit [Ping timeout: 480 seconds]
17:33rzhikharevich has joined #asahi-dev
17:54svenjannau: ack, I’m just a bit concerned about how to resume properly now with the child/parent thing
17:57svenhrm, can you wake macOS from sleep from an attached tbt keyboard?
17:58jannaudoes thunderbolt expect that the things are exactly the same after wakeup? I guess it can't really since it has to deal with hotplug during suspend anyway
17:59___nick___ has joined #asahi-dev
17:59jannauwouldn't a keyboard use usb2?
18:00jannauI'd expect that even keyboards with usb3 hub still use usb2 for the keyboard itself
18:01svenhm, true
18:01jannaumy hp dock has a (power) button but I think that's a usb2 device
18:01svenand i guess resume would first resume the parent anyway which would have to set everything up again
18:01amarioguy has quit [Ping timeout: 480 seconds]
18:01amarioguy has joined #asahi-dev
18:01svenso yeah, maybe just block resume for now
18:01svenwell, suspend and resume :-)
18:02chaos_princessnot if it is on a "true" tb hub, as that uses a pcie-attached usb controller inside the doc
18:02__nick__ has quit [Ping timeout: 480 seconds]
18:02jannauah, yes thunderbolt3 hubs
18:04svenonly tricky part might be making sure that atcphy is up and configure before i bring up acio again
18:06jannaushouldn't device_links ensure that
18:06svenwait.. is atcphy even brought down on suspend? i guess it has to, doesn't it?
18:08jannauno pm ops either so likely the explonation why usb devices need to be replugged after suspend
18:09svenlovely
18:09svenbut that should be doable then
18:11sveni'll just prevent suspend/resume as long as there's an active tbt connection then for v1
18:40sveni think prevent the suspend in the prepare callback should be even earlier
18:40sven*preventing
18:40bennofs3 has quit [Ping timeout: 480 seconds]
18:45bennofs3 has joined #asahi-dev
18:48svenbut yeah.. maybe i should add suspend/resume to atcphy first then :D
18:54mikroskeem has quit [Quit: bye]
18:56mikroskeem has joined #asahi-dev
19:00mikroskeem has quit []
19:02mikroskeem has joined #asahi-dev
19:47svenok, i think with proper atcphy suspend/resume tbt should also be fine
19:53okb has joined #asahi-dev
20:06___nick___ has quit [Remote host closed the connection]
20:27tryhard[m]chaos_princess: applied your review
21:14___nick___ has joined #asahi-dev
21:49rakyah has joined #asahi-dev
21:51lyudess has quit []
21:51Lyude has joined #asahi-dev
21:59rakyah has quit [Quit: rakyah]
22:06bennofs3 has quit [Ping timeout: 480 seconds]
22:16bennofs3 has joined #asahi-dev
22:43al3xtjames2 has quit [Quit: al3xtjames2]
22:46elvishjerricco7 has joined #asahi-dev
22:52elvishjerricco has quit [Ping timeout: 480 seconds]
22:52elvishjerricco7 is now known as elvishjerricco
23:53hardy94 has quit [Quit: hardy94]
…