#
✕ nick:sven
#asahi-dev · 2025-09-04 21:13
svenno idea why it would try to use swiotlb then :/
#asahi-dev · 2025-09-04 21:13
sveni think that's it
#asahi-dev · 2025-09-04 21:05
svenbut i don't think we marked the internal pci bus as unsafe
#asahi-dev · 2025-09-04 21:04
svene.g. anything that uses dma_map_sg to directly map buffers instead of using the coherent api to allocate buffers
#asahi-dev · 2025-09-04 21:04
sven*aligned
#asahi-dev · 2025-09-04 21:04
sveni thought it also forces anything that's not paged alined through swiotlb
#asahi-dev · 2025-09-04 20:58
sveni don't remember if that forces swiotlb
#asahi-dev · 2025-09-04 20:58
svenuh.. is the bus marked as untrusted?
#asahi-dev · 2025-09-04 11:15
svenI don’t care anymore at this point
#asahi-dev · 2025-09-04 11:07
svenwe can probably drop that once atcphy is in then
#asahi-dev · 2025-09-04 11:07
sven🙃
#asahi-dev · 2025-09-04 11:03
svenjannau: oh, yes, I didn’t realize that commit isn’t even upstream
#asahi-dev · 2025-09-04 11:03
svenbut I’ll check this weekend for the atcphy axi2af, we could drop the entire axi2af mmio range then. I only know that macos always applies them
#asahi-dev · 2025-09-04 11:02
svenmaybe it also just happens to work :D
#asahi-dev · 2025-09-04 11:01
svenjannau: I don’t think that necessary anymore, should be easy to test
#asahi-dev · 2025-09-04 11:01
svenchaos_princess: never checked, I only know that neither fuses nor the core atcphy tunable survive that (or even a atcphy reset)
#asahi-dev · 2025-09-03 14:02
svenI’ll keep chipping away at usb3 and tbt then
#asahi-dev · 2025-09-03 14:02
svenhah :D
#asahi-dev · 2025-09-03 13:57
svenwant me to work on the Feedback for hid?
#asahi-dev · 2025-09-03 13:56
sven*:)
#asahi-dev · 2025-09-03 13:56
svenaww.. after your vacation then :/
#asahi-dev · 2025-09-03 10:31
svenmaybe we'll see something if we enable xhci tracing / debug prints
#asahi-dev · 2025-09-03 09:08
svenit's not gonna stop me from submitting v2 this weekend anyway because we can always fix this later
#asahi-dev · 2025-09-03 09:08
sveni don't think I see it on macos though
#asahi-dev · 2025-09-03 09:08
svenyeah, that issue may have been there all along
#asahi-dev · 2025-09-03 09:02
svenif only now that 5 seconds delay disappeared :D
#asahi-dev · 2025-09-03 09:02
svenand it also allowed to drop the usb role from typec_mux_state
#asahi-dev · 2025-09-03 09:02
svenand this glue is also going to be easier to adjust for usb4 tunneling
#asahi-dev · 2025-09-03 09:01
svenalso managed to find out that the pipehandler lock not acked issue was usually caused because SUSPHY was in the wrong state
#asahi-dev · 2025-09-03 09:01
svenyeah, it's much nicer this way
#asahi-dev · 2025-09-03 08:00
svenand at that point atcphy should already be done
#asahi-dev · 2025-09-03 08:00
sveni don't think there's any 5s delay anywhere in atcphy
#asahi-dev · 2025-09-03 07:59
svennot yet
#asahi-dev · 2025-09-03 07:29
sven[ 82.583044] usb 4-1: new SuperSpeed USB device number 2 using xhci-hcd
#asahi-dev · 2025-09-03 07:29
sven[ 77.516615] hub 4-0:1.0: 1 port detected
#asahi-dev · 2025-09-03 07:26
svendon't see that ACIOPHY_TOP_PHY_STAT_LN0_UNK0 timeout though
#asahi-dev · 2025-09-03 07:26
svenjannau: even when forcing usb3/dp mode i still see the 5 seconds delay when directly connecting my usb3 device :/
#asahi-dev · 2025-09-02 22:14
sventhanks
#asahi-dev · 2025-09-02 22:14
svenAck
#asahi-dev · 2025-09-02 21:51
sven+hub
#asahi-dev · 2025-09-02 21:51
svencan reproduce it here as well when directly connecting the device without the usb3/dp inbetween
#asahi-dev · 2025-09-02 21:46
svenand I never understood _why_. maybe we do now
#asahi-dev · 2025-09-02 21:46
svenohh… yup
#asahi-dev · 2025-09-02 21:44
svenok, weird but at least that explains the difference
#asahi-dev · 2025-09-02 21:38
svenI’m traveling for $work for the rest of the week but I’ll compare macOS vs. my branch on the weekend to see what’s going on
#asahi-dev · 2025-09-02 21:37
svenhmmm… maybe because I‘m using a usb3/dp connected to my usb 3 hdd here
#asahi-dev · 2025-09-02 20:34
svenhrm, something else is broken then
#asahi-dev · 2025-09-02 20:27
svenI hate this thing :(
#asahi-dev · 2025-09-02 20:20
sventhat makes no sense to me but it’s a difference between Asahi tree and that branch
#asahi-dev · 2025-09-02 20:20
svencan you try moving the phy_set_mode calls to after dwc3_host_init?

plus ›