#
✕ nick:Rafay
#asahi-dev · 2026-08-14 17:46
Rafaysven: It was almost entirely m1n1 overhead. After not tracing acio like jannau said, I get 15Gbps which is the max the other host supports.
#asahi-dev · 2026-08-14 16:17
Rafayon every tbt connect.
#asahi-dev · 2026-08-14 16:16
Rafayjannau: I have an intel controller as well and I have to power off, power on and then pci rescan to get it working.
#asahi-dev · 2026-08-14 15:30
RafayOh. But macos sonoma itself is not getting those speeds. So maybe it could be a bug that asahi ends up reproducing since we can only trace sonoma?
#asahi-dev · 2026-08-14 15:26
Rafaysven: This is not part of the scope, but I noticed something really weird. MacOS Sonoma with m1n1 -> Linux is getting 1.2Gb/sec and Asahi gets 500Mb/sec but then a non-m1n1 macos Tahoe gets 19/20Gbsec? Same cable, same host on the other side. Is there a way to trace that version of mac os?
#asahi-dev · 2026-08-13 18:51
RafayYeah. It still has usable speeds.
#asahi-dev · 2026-08-13 18:49
RafayYeah. Does it work for you? I think you said it's broken for you as well?
#asahi-dev · 2026-08-13 18:48
RafayWhat about the other lane?
#asahi-dev · 2026-08-13 18:44
Rafaysven: done
#asahi-dev · 2026-08-13 17:16
RafayAround 250-400 Mbits/sec.
#asahi-dev · 2026-08-13 17:15
RafayAlso, there was some issue with the other hosts thunderbolt controller from tracing and breaking in m1n1, after reset I am getting much better speeds.
#asahi-dev · 2026-08-13 17:13
Rafaysven: 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
#asahi-dev · 2026-08-13 17:11
Rafaysven: Yeah you are right, let me double check.
#asahi-dev · 2026-08-13 17:09
RafayIt modifies the tps6598x driver only, right?
#asahi-dev · 2026-08-13 17:05
Rafaysven: [ 53.071650] thunderbolt-apple-acio 501ac0000.nhi: TBT cable: cable_mode 0x0, device_mode 0x0, enter_vdo 0x3, orientation 2 -> cable info 0x603
#asahi-dev · 2026-08-12 19:10
Rafaysven: No idea about that.
#asahi-dev · 2026-08-12 18:58
Rafaysven: Yeah, that's weird.
#asahi-dev · 2026-08-12 18:54
Rafaysven: haha. By this time you probably have more info about tbt than anyone else.
#asahi-dev · 2026-08-12 18:19
Rafaysven: Yeah. Btw, do you think this speed issue is unrelated to the tbt driver and maybe related to other stuff like the networking stack?
#asahi-dev · 2026-08-12 10:36
Rafaysven: Speeds from macos are around 1.8-2 Gbits/sec . I think it's because the other host is an older device with only thunderbolt 3 support (Intel Alpine Ridge 2c).
#asahi-dev · 2026-08-12 08:27
Rafaychaos_princess: sorry, I meant TCP.
#asahi-dev · 2026-08-12 08:26
Rafaychaos_princess: UDP
#asahi-dev · 2026-08-12 08:23
Rafayiperf output: https://pastebin.com/mVR30GkX
#asahi-dev · 2026-08-12 08:20
Rafaysven: The above line is also in my dmesg quite late after the thunderbolt messages.
#asahi-dev · 2026-08-12 08:19
Rafay[ 2862.192205] TCP: thunderbolt0: Driver has suspect GRO implementation, TCP performance may be compromised.
#asahi-dev · 2026-08-12 08:17
RafayI have no other interface, besides loopback and disconnected ethernet.
#asahi-dev · 2026-08-12 08:04
Rafayhigh_speed_lane.c:289: Gen2/3 link error. lane=1, error=82
#asahi-dev · 2026-08-12 08:04
RafayAlso aside from error 88: [ 63.138266] thunderbolt-apple-acio 501ac0000.nhi: RTKit: syslog message: high_speed_lane.c:289: Gen2/3 link error. lane=0, error=84
#asahi-dev · 2026-08-12 08:00
RafayThe iperf3 speeds are very low and it breaks by the end of the test. I don't know why it's breaking at the end but speeds are around 10.5 Mbits/s which is quite low.
#asahi-dev · 2026-08-12 00:17
Rafayhttps://pastebin.com/QS9tAabE
#asahi-dev · 2026-08-12 00:15
RafayUghh, that forcing 0x413 is breaking link init/training and therefore I am getting USB 2 speeds.
#asahi-dev · 2026-08-11 23:26
RafayThe other host is x86.
#asahi-dev · 2026-08-11 23:24
Rafaysven: I forced 0x413 in apple.c and now I am able to ping and nc.
#asahi-dev · 2026-08-11 18:32
RafayI am not running the latest commit, let me fetch that.
#asahi-dev · 2026-08-11 18:32
Rafay[ 193.553395] thunderbolt-apple-acio 501ac0000.nhi: TBT cable: cable_mode 0x0, device_mode 0x0, enter_vdo 0x3, orientation 2 -> cable info 0x603
#asahi-dev · 2026-08-11 17:18
Rafaysven: here's your tracer on my machine if it helps. Linux: https://pastebin.com/V8kfgXPH Mac: https://pastebin.com/rGZNU87Y
#asahi-dev · 2026-08-11 16:55
RafayOh, got it.
#asahi-dev · 2026-08-11 16:53
Rafaysven: Is this lspcie output expected? I think some nodes are missing. https://pastebin.com/PeSMcLTQ
#asahi-dev · 2026-08-11 16:07
Rafayt8112-j473
#asahi-dev · 2026-08-11 16:06
RafayYes.
#asahi-dev · 2026-08-11 16:06
Rafaysven: Here's the trace under linux. https://gist.github.com/rafayahmed317/d7d8768cec63178c0c15a8983af7d553 . Just to remind you, on my setup, thunderbolt_net doesn't create a device and xdomain doesn't initialize. You already have the mac trace: https://gist.github.com/rafayahmed317/e083de8e478830dc37969ca0b36750e7
#asahi-dev · 2026-08-11 09:47
RafayLemme do that.
#asahi-dev · 2026-08-11 09:44
RafayYeah.
#asahi-dev · 2026-08-11 09:41
Rafaysven: These are the logs, just tbt-reset-wip at commit 98588460927272c731c6018af25463986fdf9e0e and the patch you gave me. I didn't add the later patch you sent here and I tried with e2e=0 with no luck. https://pastebin.com/JxrwH0wK
#asahi-dev · 2026-08-10 21:56
RafayThen it's not creating the device for me. I copied the earlier config, added CONFIG_USB4_APPLE_SOC and applied the patch and compiled. The new dtb causes my wifi to stop working as well, but that could be unrelated.
#asahi-dev · 2026-08-10 21:38
RafayAh, now that I read the above messages, I think that's expected.
#asahi-dev · 2026-08-10 21:06
Rafaysven: The thunderbolt driver seems to be recognizing the device and printing out messages, but I modprobed thunderbolt-net and it didn't assign an ip or even create a device in ip link show.
#asahi-dev · 2026-08-09 18:52
RafayLemme try. Will take anywhere from a few minutes to a few hours though. First time compiling asahi.
#asahi-dev · 2026-08-09 18:19
RafayIn the adt it's written as iop,mxwrap-acio .
#asahi-dev · 2026-08-09 18:09
RafayHere's a github link if you still can't download: https://gist.github.com/rafayahmed317/e083de8e478830dc37969ca0b36750e7

plus ›