#asahi-dev · 2025-04-07 22:38
jannau0x1800170180 is within the BAR#asahi-dev · 2025-04-07 22:36
jannauL2C_ERR_STS: 0x82, L2C_ERR_ADR: 0x42c3501800170180, L2C_ERR_INF: 0x110000003#asahi-dev · 2025-04-07 22:32
jannaus/with the //#asahi-dev · 2025-04-07 22:32
jannauwith the ah, yes. it's coming back to me. and it's annoyingly async so PC doesn't make sense when it arrives#asahi-dev · 2025-04-07 22:11
jannauyes, unaligned reads and writes seem to work#asahi-dev · 2025-04-07 22:10
jannauthat is what xe triggers in linux on load#asahi-dev · 2025-04-07 22:10
jannauoh no. esr 0xbe000000 is "32-bit instruction trapped"#asahi-dev · 2025-04-07 22:08
jannauso the xe issue I still see could be still page size related#asahi-dev · 2025-04-07 22:08
jannauNormal NC seems to work after mapping it as such in m1n1 #asahi-dev · 2025-04-07 21:52
jannaubut of course that's not true under the hypervisor#asahi-dev · 2025-04-07 21:52
jannauyes, m1n1 is using Device-nGnRE. I ignored that because it doesn't matter after linux is booted#asahi-dev · 2025-04-07 21:49
jannauerr, mapped as normal memory#asahi-dev · 2025-04-07 21:49
jannaushould be mapped as device. xe driver uses iomap_wc()#asahi-dev · 2025-04-07 21:44
jannauI've pushed asahi-6.14.1-1 to github. Please test#asahi-dev · 2025-04-07 21:43
jannaufor unaligned accesses to vram BAR from m1n1's hypervisor I get unaligned SErros as one would expect#asahi-dev · 2025-04-07 21:41
jannauprobably no unaligned access to PCI BAR memory on mac pro PCIe slots. intel dg2 with 16k fixes is not particularly happy (many firmware schedular job timeouts) and I get strange SErrors#asahi-dev · 2025-04-04 09:14
jannauarch/arm64/boot/dts/qcom/sa8540p-ride.dts has a PM40028 switch but doesn't model devices behind the switch#asahi-dev · 2025-04-04 09:03
jannaunot sure how to model the reset and power enable gpios of devices behind the switch in the devicetree#asahi-dev · 2025-04-04 09:01
jannaumaz: mem assigment breaks down with hotplug but if all devices are powered up before booting it mostly works after putting "bus-range = <0 8>;" in the device tree#asahi-dev · 2025-04-04 08:58
jannauI haven't checked beyond quickly testing that old radeon hd4350#asahi-dev · 2025-04-04 08:56
jannauyes, switchtec. it's a microchip PM40100#asahi-dev · 2025-04-04 08:52
jannaupcie setup is not fully working. the switch management endpoint is not working and the devices behind the switch only come up with PCIe gen1#asahi-dev · 2025-04-04 08:51
jannauI got it refurbished for a little bit over half price. still expensive though#asahi-dev · 2025-04-04 08:28
jannaumaz: it has 2 root ports which was incompatible with the bus_notifier approach#asahi-dev · 2025-03-31 11:26
jannauthe change is referenced via Link: in the commit message#asahi-dev · 2025-03-31 11:24
jannaufeel free to add my tested-by: I'll reply in a moment#asahi-dev · 2025-03-31 11:14
jannausimilar situation as the m2 pro mac mini. When I tried to submit one of the PCI maintainers wasn't convinced it's needed at all#asahi-dev · 2025-03-31 11:12
jannaumaz: upstream "needs" it for the imac with 2 usb-c ports (t8103-j457) since commit a0189fdfb73d ("arm64: dts: apple: t8103: Disable unused PCIe ports")#asahi-dev · 2025-03-30 09:25
jannauI wonder how apple would do that. I would assume the features/regs are not be disabled for custom darwin kernels with permissive security#asahi-dev · 2025-03-30 09:23
jannauprobably defaults / everything disabled#asahi-dev · 2025-03-30 09:11
jannauare those inaccessible system regs all apple architecture extensions?#asahi-dev · 2025-03-29 09:00
jannauyes, it still boots. tested to be sure#asahi-dev · 2025-03-29 08:53
jannauI think it still boots not 100% sure I tested though#asahi-dev · 2025-03-29 08:42
jannauapalos: I did that last week: https://gist.github.com/jannau/73388b7cc9599ab6fea8fdc8ce3278ad#asahi-dev · 2025-03-29 08:05
jannaufl0_id: have you seen https://github.com/AsahiLinux/docs/issues/85 and https://github.com/bluez/bluez/issues/722#asahi-dev · 2025-03-27 22:34
jannauI'm still confused. probe() is only called once but `let dma_chan = data.request_dma_channel()?;` doesn't return. a print on the next line is seen in the log#asahi-dev · 2025-03-27 22:31
jannauthe audio device is created before SndSocAopDriver::new(). that's the last call in probe. not to confuse with SndSocAopData::new() which is called earlier#asahi-dev · 2025-03-27 22:25
jannauno#asahi-dev · 2025-03-27 22:24
jannaufrom_err_ptr(ERR_PTR(-PROBE_DEFER)).is_err() is false?#asahi-dev · 2025-03-27 22:10
jannauyes#asahi-dev · 2025-03-27 22:02
jannauit's only called once#asahi-dev · 2025-03-27 22:00
jannaudo you expect it to be called more than once? I'd be surprised if it did#asahi-dev · 2025-03-27 21:51
jannaureproduced without pdm syslog messages: https://paste.centos.org/view/1ac5e5a1#asahi-dev · 2025-03-27 21:37
jannauyes, unable to get dma channel. failed to attach devicem and than probe failed with -5#asahi-dev · 2025-03-27 21:35
jannaufasiled to get dma channel, than aop syslog pdm errors. let me see if I can reproduce it#asahi-dev · 2025-03-27 21:34
jannauthe platform/of probe code uses some properties like iommus, dmas, ... for probe ordering. I've worked with https://paste.centos.org/view/8db1f78f around for now#asahi-dev · 2025-03-27 21:29
jannaunot sure how the error is propagated. it ended up with returning -EIO#asahi-dev · 2025-03-27 21:26
jannauissue is likely that nothing enforces the order from the "dmas" for the manual platformdevice creation#asahi-dev · 2025-03-27 21:25
jannauchaos_princess: have you seen a probe order issue between aop_audio and aop_admac? Not sure if I were just lucky to hit that in 6.14 or if it's a 6.14 driver-core/platform probe change#asahi-dev · 2025-03-26 11:51
jannauchadmed: can https://gitlab.freedesktop.org/pipewire/wireplumber/-/merge_requests/693 and https://gitlab.freedesktop.org/pipewire/pipewire/-/issues/4573 be closed?