#
…

2026-08-13

00:00voidah has quit [Ping timeout: 480 seconds]
00:08hdbngr has quit [Remote host closed the connection]
00:08voidah has joined #asahi-dev
00:21voidah has quit [Ping timeout: 480 seconds]
00:24hdbngr has joined #asahi-dev
00:28voidah has joined #asahi-dev
00:33hdbngr has quit [Ping timeout: 480 seconds]
00:37voidah has quit [Ping timeout: 480 seconds]
00:41hdbngr has joined #asahi-dev
00:45voidah has joined #asahi-dev
00:48leah has quit [Quit: i will return]
00:50hdbngr has quit [Ping timeout: 480 seconds]
00:54voidah has quit [Ping timeout: 480 seconds]
00:58hdbngr has joined #asahi-dev
01:02voidah has joined #asahi-dev
01:06hdbngr has quit [Ping timeout: 480 seconds]
01:10voidah has quit [Ping timeout: 480 seconds]
01:13Acelogic_ has joined #asahi-dev
01:13Acelogic_ has quit []
01:13Acelogic_ has joined #asahi-dev
01:16hdbngr has joined #asahi-dev
01:18voidah has joined #asahi-dev
01:24hdbngr has quit [Ping timeout: 480 seconds]
01:26voidah has quit [Ping timeout: 480 seconds]
01:32hdbngr has joined #asahi-dev
01:35JinketomyMasheldia[m] has joined #asahi-dev
01:37__nick__ has quit []
01:39___nick___ has joined #asahi-dev
01:40hdbngr has quit [Ping timeout: 480 seconds]
01:40___nick___ has quit []
01:43___nick___ has joined #asahi-dev
01:48hdbngr has joined #asahi-dev
01:57hdbngr has quit [Ping timeout: 480 seconds]
02:00yurix has quit [Remote host closed the connection]
02:00yurix has joined #asahi-dev
02:04hdbngr has joined #asahi-dev
02:12hdbngr has quit [Ping timeout: 480 seconds]
02:16bgb has quit [Remote host closed the connection]
02:20nnym has quit [Ping timeout: 480 seconds]
02:20pb17 has quit [Ping timeout: 480 seconds]
02:22hdbngr has joined #asahi-dev
02:22bgb has joined #asahi-dev
02:24chadmed has quit [Quit: Konversation terminated!]
02:24chadmed has joined #asahi-dev
02:28pb17 has joined #asahi-dev
02:29nnym has joined #asahi-dev
02:31hdbngr has quit [Ping timeout: 480 seconds]
02:37hdbngr has joined #asahi-dev
02:37nnym has quit [Ping timeout: 480 seconds]
02:45nnym has joined #asahi-dev
02:46hdbngr has quit [Ping timeout: 480 seconds]
02:47tobhe_ has joined #asahi-dev
02:53nnym has quit [Ping timeout: 480 seconds]
02:54hdbngr has joined #asahi-dev
02:54tobhe has quit [Ping timeout: 480 seconds]
03:01nnym has joined #asahi-dev
03:02hdbngr has quit [Ping timeout: 480 seconds]
03:10nnym has quit [Ping timeout: 480 seconds]
03:11hdbngr has joined #asahi-dev
03:16bgb has quit [Remote host closed the connection]
03:18nnym has joined #asahi-dev
03:19bgb has joined #asahi-dev
03:20hdbngr has quit [Ping timeout: 480 seconds]
03:26nnym has quit [Ping timeout: 480 seconds]
03:27hdbngr has joined #asahi-dev
03:30myria has quit [Ping timeout: 480 seconds]
03:35nnym has joined #asahi-dev
03:36hdbngr has quit [Ping timeout: 480 seconds]
03:43nnym has quit [Ping timeout: 480 seconds]
03:44hdbngr has joined #asahi-dev
03:52hdbngr has quit [Ping timeout: 480 seconds]
03:53nnym has joined #asahi-dev
03:54nora has joined #asahi-dev
03:56nora_ has quit [Ping timeout: 480 seconds]
04:01hdbngr has joined #asahi-dev
04:01nnym has quit [Ping timeout: 480 seconds]
04:09nnym has joined #asahi-dev
04:10hdbngr has quit [Ping timeout: 480 seconds]
04:17nnym has quit [Ping timeout: 480 seconds]
04:17hdbngr has joined #asahi-dev
04:25nnym has joined #asahi-dev
04:26hdbngr has quit [Ping timeout: 480 seconds]
04:34hdbngr has joined #asahi-dev
04:43hdbngr has quit [Ping timeout: 480 seconds]
04:49croissant has joined #asahi-dev
04:50myria has joined #asahi-dev
04:51hdbngr has joined #asahi-dev
04:54croissant` has quit [Ping timeout: 480 seconds]
05:08hdbngr has quit [Ping timeout: 480 seconds]
05:09croissant has quit [Quit: Leaving]
05:09croissant has joined #asahi-dev
05:14hdbngr has joined #asahi-dev
05:23hdbngr has quit [Ping timeout: 480 seconds]
05:30hdbngr has joined #asahi-dev
05:40Carlos has quit [Quit: Carlos]
05:43hdbngr has quit [Ping timeout: 480 seconds]
05:51hdbngr has joined #asahi-dev
05:53lion328 has quit [Remote host closed the connection]
05:55lion328 has joined #asahi-dev
06:00hdbngr has quit [Ping timeout: 480 seconds]
06:07hdbngr has joined #asahi-dev
06:16hdbngr has quit [Ping timeout: 480 seconds]
06:25hdbngr has joined #asahi-dev
06:27bennofs3 has quit [Ping timeout: 480 seconds]
06:31bennofs3 has joined #asahi-dev
06:36pb17 has quit [Ping timeout: 480 seconds]
06:37pb17 has joined #asahi-dev
06:40hdbngr has quit [Ping timeout: 480 seconds]
06:43myria has quit [Remote host closed the connection]
06:43hardy94 has joined #asahi-dev
06:43myria has joined #asahi-dev
06:49hdbngr has joined #asahi-dev
06:57hdbngr has quit [Ping timeout: 480 seconds]
07:05hdbngr has joined #asahi-dev
07:14hdbngr has quit [Ping timeout: 480 seconds]
07:22hdbngr has joined #asahi-dev
07:30hdbngr has quit [Ping timeout: 480 seconds]
07:35Acelogic_ has quit []
07:35hdbngr has joined #asahi-dev
07:35Acelogic_ has joined #asahi-dev
07:36bgb has quit [Remote host closed the connection]
07:44hdbngr has quit [Ping timeout: 480 seconds]
07:51Acelogic_ has quit []
07:51Acelogic_ has joined #asahi-dev
07:52bgb has joined #asahi-dev
07:52hdbngr has joined #asahi-dev
07:56Acelogic_ has quit []
07:56Acelogic_ has joined #asahi-dev
08:01hdbngr has quit [Ping timeout: 480 seconds]
08:08hdbngr has joined #asahi-dev
08:14Acelogic_ has quit []
08:14Acelogic_ has joined #asahi-dev
08:17hdbngr has quit [Ping timeout: 480 seconds]
08:18Acelogic_ has quit []
08:18Acelogic_ has joined #asahi-dev
08:25hdbngr has joined #asahi-dev
08:34hdbngr has quit [Ping timeout: 480 seconds]
08:35picnoir has quit [Ping timeout: 480 seconds]
08:36hdbngr has joined #asahi-dev
08:44picnoir has joined #asahi-dev
08:45hdbngr has quit [Ping timeout: 480 seconds]
08:52hdbngr has joined #asahi-dev
08:55hardy94 has quit [Quit: hardy94]
09:01hdbngr has quit [Ping timeout: 480 seconds]
09:09hdbngr has joined #asahi-dev
09:23hdbngr has quit [Ping timeout: 480 seconds]
09:32hdbngr has joined #asahi-dev
09:45hdbngr has quit [Ping timeout: 480 seconds]
09:55hdbngr has joined #asahi-dev
10:03hdbngr has quit [Ping timeout: 480 seconds]
10:08leah has joined #asahi-dev
10:09leah has quit []
10:09leah has joined #asahi-dev
10:11hdbngr has joined #asahi-dev
10:15myria has quit [Remote host closed the connection]
10:17myria has joined #asahi-dev
10:37hdbngr has quit [Ping timeout: 480 seconds]
10:38hdbngr has joined #asahi-dev
10:47hdbngr has quit [Ping timeout: 480 seconds]
10:50jannauI've tagged asahi-7.1.8-1
10:54hdbngr has joined #asahi-dev
10:56chadmedi might be able to get most of the remaining audio stuff upstream by 7.4 or 7.5
10:57chadmedactually 7.4 is unrealistic
10:57chadmedthe i2s stuff that got missed last time will be in 7.3 so that pretty much covers all the non-macaudio prerequisites
11:03hdbngr has quit [Ping timeout: 480 seconds]
11:03chadmedi need to see if the gpio-shared-proxy thing will work for us. looks like it should
11:07chadmed"Selected by: ARCH_QCOM"
11:10hdbngr has joined #asahi-dev
11:17sofusim very confused by the interchange thing, it seems to just unset a flag and set output addresses to zero
11:17sofuschadmed: if i send you a hexdump would you be able to tell if its an interchange thing?
11:18chadmedmaybe? i dont know what id be looking at or for though :P
11:18chadmedother than hope and pray that theres a fourcc in there that matches a dcp one, but i doubt it
11:18jannauthat's nothing strange. that just says that ARCH_QCOM has a "select GPIO_SHARED_PROXY". macaudio or ARCH_APPLE would do the same if it is usable
11:19hdbngr has quit [Ping timeout: 480 seconds]
11:19chadmedyeah ive already sorted it, it was just something i didnt expect reading some of the commit descriptions
11:19sofuschadmed: no like an raw buffer, just data
11:20chadmedrender it into linear space, if its mostly black/null with some coloured specks in there then it's a compressed fb
11:20chadmedyou might even be able to see a tiling pattern
11:25chadmed has quit [Quit: Konversation terminated!]
11:27hdbngr has joined #asahi-dev
11:28pb17 has quit [Ping timeout: 480 seconds]
11:29chadmed has joined #asahi-dev
11:30chadmedhmm it doesnt quite work right on suspend/resume
11:31chadmedthe i2c bus is unhappy
11:32chadmed has quit []
11:36hdbngr has quit [Ping timeout: 480 seconds]
11:39hdbngr has joined #asahi-dev
11:44pb17 has joined #asahi-dev
11:48hdbngr has quit [Ping timeout: 480 seconds]
11:55hdbngr has joined #asahi-dev
12:03hdbngr has quit [Ping timeout: 480 seconds]
12:08chadmed has joined #asahi-dev
12:08chadmedsigh all consumers now have to agree on both assert and deassert
12:12hdbngr has joined #asahi-dev
12:20hdbngr has quit [Ping timeout: 480 seconds]
12:28hdbngr has joined #asahi-dev
12:36hdbngr has quit [Ping timeout: 480 seconds]
12:40hdbngr has joined #asahi-dev
12:41chadmed has quit [Quit: Konversation terminated!]
12:42chadmed has joined #asahi-dev
12:49hdbngr has quit [Ping timeout: 480 seconds]
12:57hdbngr has joined #asahi-dev
13:06hdbngr has quit [Ping timeout: 480 seconds]
13:15hdbngr has joined #asahi-dev
13:23hdbngr has quit [Ping timeout: 480 seconds]
13:31hdbngr has joined #asahi-dev
13:41MikeM64 has quit [Ping timeout: 480 seconds]
13:49hdbngr has quit [Ping timeout: 480 seconds]
13:57hdbngr has joined #asahi-dev
14:02svenif it doesn't work i'd like to see that "TBT cable: ..." message again
14:05hdbngr has quit [Ping timeout: 480 seconds]
14:14hdbngr has joined #asahi-dev
14:23hdbngr has quit [Ping timeout: 480 seconds]
14:31hdbngr has joined #asahi-dev
14:40hdbngr has quit [Ping timeout: 480 seconds]
14:41hdbngr has joined #asahi-dev
14:53hdbngr has quit [Ping timeout: 480 seconds]
14:56qyliss has quit [Quit: bye]
14:56qyliss has joined #asahi-dev
15:02hdbngr has joined #asahi-dev
15:11hdbngr has quit [Ping timeout: 480 seconds]
15:18hdbngr has joined #asahi-dev
15:23mripard has quit [Remote host closed the connection]
15:24svenchaos_princess: can you run tcpdump on the macos side and see if the ping packets arrive? i'd like to figure out if the packets already disappear on the way there or on the way back
15:25chaos_princesssure, will do in a bit.
15:26chaos_princessanother thing i noticed is that i again get link errors (the same kind that are from wrong vse) if i connect m1 <-> m1pro, while m1pro <-> m3max works
15:26svenhm... weird
15:26chaos_princessbase m1 technically does not list tb4 only "usb4/tb", while the other two do
15:27hdbngr has quit [Ping timeout: 480 seconds]
15:27hdbngr has joined #asahi-dev
15:27sveniirc the differnce between usb4/tb and tb4 is just all the optional features required and at least 2 tunneled display streams
15:27svendoes the base m2 list tb4? because that one works as well it seems
15:28chaos_princessthat one does
15:28mripard has joined #asahi-dev
15:30svenwe could also try kicking it from the hv, which acio do you test again? acio1?
15:31chaos_princessi can do whatever
15:32mripard_ has joined #asahi-dev
15:33svenokay for acio1 something like this i think
15:33sventx = read32(0xb01f14010)
15:33svenwrite32(0xb01f14010, tx & ~0x80000000); write32(0xb01f84010, rx & ~0x80000000)
15:33svenrx = read32(0xb01f84010)
15:33svenwrite32(0xb01f14010, tx); write32(0xb01f84010, rx)
15:34chaos_princesswhat's that supposed to do?
15:35mripard__ has joined #asahi-dev
15:35svenclears and then sets the enable bit for ring1
15:35svenit might just break everything, but given that it's already not working it can't hurt
15:36hdbngr has quit [Ping timeout: 480 seconds]
15:36sveni'm wondering if it's some weird "this controller only updates state when enable is set" or something similar
15:37svenbecause the traces look very very similar, it's just that one them doesn't work.
15:38mripard has quit [Ping timeout: 480 seconds]
15:39hdbngr has joined #asahi-dev
15:41mripard_ has quit [Ping timeout: 480 seconds]
15:48chaos_princessso, wireshark does not see anything
15:48chaos_princesswhat may also be interesting is that the error i get on the linux side is "destination host unreachable"
15:49chaos_princessoh, that's because it can't arp
15:51svenwhat if you try the other way around? so pings from macos and wireshark on linux?
15:51hdbngr has quit [Ping timeout: 480 seconds]
15:51chaos_princessdoing the hypervisor kick did not help
15:51chaos_princessthat is how i figured out that it can't arp, looked at what macos does when sending a ping, and it ended up sending arp requests to nowhere
15:52chaos_princessdo we want to try breaking lane bonding again?
15:53svenoh right, that didn't work out because my first attempt was 5 minutes before i went to bed :D
15:53svenyeah, let's see how to break that without breaking the entire connection :D
15:53svenlast attempt was -ENODEV at the beginning of something in xdomain.c, wasnt it?
15:55chaos_princessthe caller is ignoring the return value there
15:56MikeM64 has joined #asahi-dev
15:56svenhrm... but didn't something fail with -ENODEV then?
15:58chaos_princessno, it got into a connect/disconnect loop
15:58svenhrm, okay. so maybe it has to be disabled much earlier. let me try to find where it decides to lane bonding
15:58sventb_xdomain_lane_bonding_enable is probably too late alreayd
15:58chaos_princesstlw = target link width
15:59chaos_princessoh, this is quite different
16:00bgb has quit [Remote host closed the connection]
16:00bgb has joined #asahi-dev
16:00hdbngr has joined #asahi-dev
16:00svenhttps://elixir.bootlin.com/linux/v7.1.8/source/drivers/thunderbolt/xdomain.c#L1619 or maybe force the else branch there to skip the entire state machine
16:01svenyeah, that's one state before it reaches your line i think
16:02svener wait.. we're looking at different sides, aren't we?. let me check which side actually initiates the bonding
16:02chaos_princessdepends on machine uuids, i think in my case it is the linux side
16:03svenokay, then forcing that else branch should work to skip the states that would eventually initiate bonding
16:03svenor maybe just do both ends to be safe :D
16:04nicolas17 has quit [Ping timeout: 480 seconds]
16:06mripard__ has quit [Remote host closed the connection]
16:09svenhuh, your mac also has a real uuid, the linux driver currently has a fallback one.
16:09hdbngr has quit [Ping timeout: 480 seconds]
16:09svenguess i should also figure out where that actually comes from
16:10svenand in the trace it's also trying to initiate that link state change request. so maybe really do both sides to be safe
16:14chaos_princesshow do i stop the other side
16:16hdbngr has joined #asahi-dev
16:18rhysmdnz has quit [Quit: Bridge terminating on SIGTERM]
16:18Guest15533 has quit [Quit: Bridge terminating on SIGTERM]
16:19sventhought you already found it, just set slw and tlw in that LINK_STATE_STATUS_REQUEST to 1
16:21chaos_princessyea, also just realized that i can do that
16:23mripard has joined #asahi-dev
16:25hdbngr has quit [Ping timeout: 480 seconds]
16:26pb17 has quit [Ping timeout: 480 seconds]
16:26chaos_princessdoes not work
16:26chaos_princessdisabling both results in disconnect/reconnect spam, disabling only slw - same as before
16:26svencan you grab a trace again?
16:27svenwith only slw
16:27sveni'm essentially out of ideas at this point :(
16:28chaos_princesstunables? fuses?
16:29svenmust be somewhere outside of acio then, because that MMIO traces contains those and iirc i compared them and they looked the same
16:29chaos_princessi mean phy-side
16:30svenyeah, worth a look
16:34hdbngr has joined #asahi-dev
16:39rhysmdnz has joined #asahi-dev
16:39pb17 has joined #asahi-dev
16:40Guest15587 has joined #asahi-dev
16:41chaos_princessis the fact that no new lines appear on tracer when i run ping normal?
16:44svenyeah, i think the tracer only looks at ring0
16:45svenwas that run just with slw=1 or also with that forced else path?
16:46chaos_princessforced else is a reconnect spam
16:46svenoh, weird
16:48svenmaybe set it to false here already https://elixir.bootlin.com/linux/v7.1.8/source/drivers/thunderbolt/xdomain.c#L1940 though i don't quite see how that should make a difference
16:51svenand while you're at it maybe also trace atcphy both linux and macos to see if i messed up fuses or tunables
16:54chaos_princessno reconnect spam, but still no packets
16:54svencan you grab a trace to confirm nothing bonded this time? maybe together with that atcphy trace
17:00sveni just realized we wouldn't see interrupts happening because the tracer has verbose=1
17:01___nick___ has quit [Ping timeout: 480 seconds]
17:02svenah, but /proc/interrupts showed they never fired so nvm
17:05Rafaysven: [ 53.071650] thunderbolt-apple-acio 501ac0000.nhi: TBT cable: cable_mode 0x0, device_mode 0x0, enter_vdo 0x3, orientation 2 -> cable info 0x603
17:05svenare you sure you applied that patch?
17:07sveni don't see how enter_vdo can every contain anything in the lowest 16bits after that one
17:08svenchaos_princess: latest trace looks like an unbonded link, so it's not lane bonding at fault :(
17:08svencan you also grab an atcphy trace from macos?
17:08chaos_princessworking on it
17:09RafayIt modifies the tps6598x driver only, right?
17:09chaos_princessidk what is up with macos under the tracer, it is just dog slow
17:10svenRafay: tbt_data.enter_vdo = (u32)le16_to_cpu(st->intel_vid_status.enter_vdo) << 16; <-- with that enter_vdo 0x3 cannot be possible
17:11sveni'm about to grab dinner, so mac os can take its time booting and getting tbt up and running :D
17:11Rafaysven: Yeah you are right, let me double check.
17:12hdbngr has quit [Ping timeout: 480 seconds]
17:13Rafaysven: Sorry, it was not installed into the correct location. [ 53.289188] thunderbolt-apple-acio 501ac0000.nhi: TBT cable: cable_mode 0x30001, device_mode 0x10001, enter_vdo 0x30000, orientation 2 -> cable info 0x413
17:14svenlooks much better
17:15RafayAlso, there was some issue with the other hosts thunderbolt controller from tracing and breaking in m1n1, after reset I am getting much better speeds.
17:16RafayAround 250-400 Mbits/sec.
17:20hdbngr has joined #asahi-dev
17:29hdbngr has quit [Ping timeout: 480 seconds]
17:36hdbngr has joined #asahi-dev
17:45hdbngr has quit [Ping timeout: 480 seconds]
17:53hdbngr has joined #asahi-dev
18:00chaos_princessjannau: you have a j314c too, right? Could you also test whether usb4net works? I am having suspicions of the kind i don't want to have
18:01hdbngr has quit [Ping timeout: 480 seconds]
18:03jannauyes, I can test later. not sure what that suspicion would be assuming you have tested that m1/linux to m1max/macos usb4net works
18:04svenfabric firewall that's only enabled on his machine in raw boot mode? :P
18:04chaos_princessno, a "is my cable just broken"
18:05chaos_princesssince it just failed to link between m1 macos and m3max macos
18:06jannauok, shall I test the same commit I tested for M1<->M1 or have there been fixes I need to take?
18:07sveni don't think any of the "fixes" we tried did anything
18:07jannauok. I'll test after dinner
18:09hdbngr has joined #asahi-dev
18:10riker77 has quit [Quit: Quitting IRC - gone for good...]
18:14rany has quit [Remote host closed the connection]
18:14rany has joined #asahi-dev
18:18hdbngr has quit [Ping timeout: 480 seconds]
18:18svenhttps://lore.kernel.org/asahi/20260813-b4-tipd-vdo-fix-v1-1-70317f2cd554@kernel.org/T/#u tipd fix, if you wanna send you tested-by Rafay. i removed one redundant line from the version you tested fwiw
18:21nicolas17 has joined #asahi-dev
18:25hdbngr has joined #asahi-dev
18:34hdbngr has quit [Ping timeout: 480 seconds]
18:36riker77 has joined #asahi-dev
18:38hdbngr has joined #asahi-dev
18:39vimproved has quit [Remote host closed the connection]
18:40vimproved has joined #asahi-dev
18:44Rafaysven: done
18:45svennow there's that m1 pro mystery left and then cleaning all this up. and dt-bindings ofc...
18:48RafayWhat about the other lane?
18:49svenoh right, that was broken for you as well
18:49svenit just wasn't the root cause of the m1 pro issue
18:49RafayYeah. Does it work for you? I think you said it's broken for you as well?
18:50sveni thought i saw in one trace that it failed to negotiate lane bonding and i also saw some syslog messages
18:50svenbut will have to check again
18:50svenas long as it kinda-sorta-works i'm happy to send a v1 even without lane bonding though
18:51RafayYeah. It still has usable speeds.
18:56hdbngr has quit [Ping timeout: 480 seconds]
19:05hdbngr has joined #asahi-dev
19:11jannauboth tbt_net and _stream appear to be broken for t8013/t6001. I did not test tbt_stream for t8103/t8103 but would expect it to work
19:13hdbngr has quit [Ping timeout: 480 seconds]
19:14svencan you compare the atcphy bringup that macos does vs. what we do?
19:15sveni'm really running out of ideas at this point
19:15svenand i assume you also see no packets in tcpdump on either side?
19:19jannauping reports "Destination unreachable: Address unreachable" / "sendmsg: No route to host" so I'd assume no
19:20svenare you still in the hv on the t6001?
19:20svenif yes, which acio port are you using?
19:20jannauthe thunderbolt0 devices have the same MAC but that doesn't seem to be an issue (was the same for t8103/t8103)
19:21jannauyes, still in the hv on both systems, using acio1 on both
19:21hdbngr has joined #asahi-dev
19:22svenread32(0xb01f10008)
19:22svenread32(0xb01f14008)
19:22svenread32(0xb01fd0000)
19:22svenplease :)
19:22svenread32(0xb01f84008)
19:22svenon the t6001
19:23jannauswitched to acio2 on the t6001 without change. I'll switch back to acio1 for the hv read32
19:23nicolas17how does thunderbolt networking work anyway? is it IP over Thunderbolt or is there an Ethernet frame layer in between too?
19:24jannauit's direct host to host network without ethernet. so the MAC address shouldn't matter
19:24svenit's essentially ethernet frames with an additional header iirc
19:25chaos_princessit kinda does, there is arp involved
19:25svenbut we should see interrupts on ring1 and chaos' /proc/interrupts showed there was nothing there
19:26svenif we saw interrupts then the packets would arrive and something else would be broken
19:27svenlast one confirms there's no interrupt pending so the issue isn't a wrong irq number in the dt
19:29svenfirst one is txring0 and PROD=CONS=1 so it's working
19:29nicolas17I do wish I could do a direct wired connection between my two laptops, guess I'd have to buy a second USBC-Ethernet dongle and use an Ethernet cable :P
19:29svennext one shows the acio never fetched even a single entry from the txring even though the kernel sent 0xe in there :/
19:30sventhird one shows the kernel prepared 0xff buffers and uhh.. the hardware sent at most one
19:30svenwhich is weird
19:30hdbngr has quit [Ping timeout: 480 seconds]
19:32jannauthis was ping from t8103, after ping on t6001 0xb01f14008 changed https://paste.debian.net/hidden/ad8b1642
19:33svenyeah, because the kernel put more stuff in there for the hardware to ignore :(
19:34svencan you do iova = read32(0xb01f84004) << 32 | read32(0xb01f84000) ... ah crap, do we have something to read a mapped buffer through dart?
19:34jannaudart.ioread()
19:35svenah, then that please for the dart :D
19:35jannauwill take a moment, I've shut down to test t6001/t6002
19:35sveni wanna see if there's actually a single entry in that ring due to that here >>> read32(0xb01f84008)
19:35sven0x100ff
19:36sveni would've expected 0xff there if nothing ever arrived
19:36svenwaaaait. can you hack the dart driver to enable all streams so that we see faults?
19:37nicolas17btw maybe you can try UDP (netcat -u) instead of ping? could make it easier to tell if the connection is working in one direction and the packet arrives at the dest but there's no reply
19:37nicolas17(it sounds like you know you're not there yet, I mean for later)
19:37sveni expect that the moment we fix this stupid crap it's just gonna work like m1/m1
19:37svenand i can see from those hw regs if something arrived or was picked up by acio
19:37chaos_princesssven: worth a try but unlikely. as a part of my tests i added stream 15 to nhi, and got no faults
19:38svenchaos_princess: ah.. :(
19:38chaos_princessbut thats just 15
19:38hdbngr has joined #asahi-dev
19:38svenlet's just enable every single stream
19:39sveni'd kinda also expect acio to just panic if DART doesn't allow a write it tires
19:39sven*tries
19:39chaos_princessand yes, my previous cable was in fact, broken(?), but the new one does not fix the problem. But at least it manages to link up more stuff.
19:43jannaut6001/t6002 doesn't work either
19:43svenas expected :(
19:49jannauhow much data do you want from the buffer? https://paste.debian.net/hidden/1b5f9711 the restlooks like it repeats more or less
19:51jannauI'm pretty sure we would get fault irqs on the dart if acio would try use a disabled stream
19:51chaos_princessi don't think so
19:51chaos_princessat least on t6000
19:51sventhat's more than enough, each entry is 16 bytes and i only need the first one
19:51svenand iirc dart only faults on enabled streams
19:54hdbngr has quit [Ping timeout: 480 seconds]
19:56svenjust double checked i got the correct dart and stream with the adt. also no issue there :(
19:56svenwait. can you specifically enable stream 0 as well?
19:57svenor do iommus = <&DIE_NODE(usb4_1_dart) 0>, iommus = <&DIE_NODE(usb4_1_dart) 1>; in the dt?
20:01jannauI did enable all stream in the hv and did not see fault
20:03svenso the t600x ADT has sids = 32771 (0x8003, so streams 0,1 and 15) and fucking remap = 258 which is 0x0102
20:03hdbngr has joined #asahi-dev
20:05chaos_princesswhat is remap about?
20:06jannauany idea what sat-dtf is?
20:07svennope, just noticed that as well
20:08bgb has quit [Remote host closed the connection]
20:08bgb has joined #asahi-dev
20:08svenchaos_princess: i vaguely remember that t8103 had a feature somewhere where you could map the hardware stream id to the stream id you actually program
20:08chaos_princessthats not confusing at all
20:09chaos_princessso, do a max verbose tracer on acio dart?
20:11svenit's usually 0102030405060708090a0b0c0d0e0f iirc
20:11hdbngr has quit [Ping timeout: 480 seconds]
20:12svenmax verbose trace on the dart and also iommus = <.. 0>, <.. 1>; would be nice
20:13chaos_princessmax verbose on macos, no?
20:13jannaum2 dart-acio1 has "remap = 258" as well
20:14jannauiommus = <.. 0>, <.. 1> did not change anything on linux
20:15svenjannau: huh, did i look in the wrong device tree then? i thought base M2 also had no remap
20:17jannausven: my fault, I looked at the m2 pro adt
20:19hdbngr has joined #asahi-dev
20:22jannaunot sure if it matters. m2 dart-dcp has remap = 0x50700000506 and I'm not aware that we're doing anything with it
20:22svenisn't the dcp dart locked?
20:23jannauyes, so we won't reset it at least
20:23svenyes, so it'll just stay however iboot programmed it
20:24jannauwhat about dart-usb* with remap = 256?
20:25svenhm... okay, that's odd
20:25svenso maybe this is just another dead end :(
20:25svenwould still be good to trace the dart with verbose=3 and just hack the linux driver to reproduce as much of that as possible to rule it out
20:28hdbngr has quit [Ping timeout: 480 seconds]
20:28jannaum1 has a much more complicated remap in dart-acio* as well
20:35jannaudo you have a way to trigger to start the dart tracer when it's powered on
20:35hdbngr has joined #asahi-dev
20:38svenhm.. right, we'd need to capture the early setup
20:38svenmaybe hook that pmgr register? though i don't know how quickly the dart comes up after that
20:39svenotherwise on the first acio register access
20:39jannauI guess we can just dump it after it is running
20:39sventrue, that also works
20:52jannauREMAP[] is [0x3010100, 0x7060504, 0xb0a0908, 0xf0e0d0c]
20:53hdbngr has quit [Ping timeout: 480 seconds]
20:53jannaudart-dcp with remap = 5 is [0x3020100, 0x7060004, 0xb0a0908, 0xf0e0d0c]
20:54jannaudart-acio1 stream 0 is enabled but has nothing mapped
20:54svenSo that’s sid 1 and 2 both mapped to 1
20:55svencan you try those two in the fdt?
20:57jannauping works
20:57sven🥳🥳🥳
20:58svenfuck you Apple, was that really necessary :D
21:00chaos_princesswhy is that feature even a thing
21:00Sobek[m]What feature ?
21:00chaos_princessthe make-darts-more-confusing one
21:00hdbngr has joined #asahi-dev
21:00svenI‘m kinda glad I obsessively tired to figure out how dart works back when m1 was released and still remembered this
21:01sven*tried
21:01chaos_princessbtw, since this is a thing, can we use it for linked darts on isp and such
21:02chaos_princessor do we want to not bother
21:02svenit would require an additional device tree property
21:02svenand it might be hard to argue for that when adding multiple iommus already works
21:03chaos_princessminor memory savings
21:04svenI don’t care tbh, I’m just not going to take part in that discussion
21:05jannauusb3 appears to be broken after thunderbolt was active
21:05jannauon a specific port
21:05sventhat’s surprising since we go through a full phy reset there
21:06chaos_princessconfirming tb link here too
21:06chaos_princess[ 5] 0.00-10.00 sec 19.9 GBytes 17.1 Gbits/sec
21:08chaos_princess[ 5] 0.00-10.00 sec 23.2 GBytes 19.9 Gbits/sec
21:08Yushi has quit [Remote host closed the connection]
21:08chaos_princessfirst is a "usb3" rated cable, the second is "usb4 gen 3"
21:09sventhat’s with bonded lanes, right?
21:09hdbngr has quit [Ping timeout: 480 seconds]
21:09Yushi has joined #asahi-dev
21:10chaos_princessi don't think so?
21:10chaos_princessreplying with target link width set to 0x1
21:10chaos_princessit is with whatever got negotiated, but without the code to manually drop the lane width
21:12chaos_princessreceive is a lot worse
21:12chaos_princess[ 5] 0.00-10.00 sec 8.67 GBytes 7.45 Gbits/sec
21:14jannauthe latter is what I see with with linux/linux
21:18hdbngr has joined #asahi-dev
21:19Lyude has quit [Quit: Bouncer restarting]
21:19jannauusb3 issue might be caused by "Pipehandler lock not acked." / "Failed to lock pipehandler" after a tbt cable disconnect
21:20Lyude has joined #asahi-dev
21:27hdbngr has quit [Ping timeout: 480 seconds]
21:32jannauno thunderbolt connection with an intel notebook
21:34jannauworks with macos but ucsi_acpi complains about duplicate partner altmodeSVID 0x8087
21:34hdbngr has joined #asahi-dev
21:44hdbngr has quit [Remote host closed the connection]
21:44hdbngr has joined #asahi-dev
21:53hdbngr has quit [Ping timeout: 480 seconds]
21:59hdbngr has joined #asahi-dev
22:07hdbngr has quit [Ping timeout: 480 seconds]
22:17hdbngr has joined #asahi-dev
22:25hdbngr has quit [Ping timeout: 480 seconds]
22:34hdbngr has joined #asahi-dev
22:48hdbngr has quit [Ping timeout: 480 seconds]
22:56hdbngr has joined #asahi-dev
23:12hdbngr has quit [Ping timeout: 480 seconds]
23:20hdbngr has joined #asahi-dev
23:32hdbngr has quit [Ping timeout: 480 seconds]
23:39hdbngr has joined #asahi-dev
23:48hdbngr has quit [Ping timeout: 480 seconds]
23:50hdbngr has joined #asahi-dev
23:59hdbngr has quit [Ping timeout: 480 seconds]
…