#asahi · 2021-06-05 09:22
jannauworks but is quite noisy and slow. should be handled in m1n1 with the ability to ignore uninteresting event types and numbers#asahi · 2021-06-05 09:08
jannauI'll try#asahi · 2021-06-05 06:31
jannauyes, it is the missing P_GET_SIMD_STATE command. I should have looked at the log first#asahi · 2021-06-05 06:25
jannauamw: sound like a regression of making checksums optional. I'm not sure if we care as long as chainloading a newer m1n1 works#asahi · 2021-06-04 23:19
jannaufor me as well, good night#asahi · 2021-06-04 23:12
jannauif 3 markers are enough we could M1racles. user space sets 1,2 or 3 and the hypervisor inserts the marker and clears the bits#asahi · 2021-06-04 23:10
jannauwe could add a virtual uart which inserts markers into mmiotrace. better than matching mmiotrace and serial out with timestamps#asahi · 2021-06-04 22:54
jannausven: https://gist.github.com/jannau/8284b9e7026c625496e1cf5263781eee connect, disconnect, connct#asahi · 2021-06-04 22:43
jannauyes, I removed the port 1 from setup_adt()#asahi · 2021-06-04 22:37
jannaumac os breaks rebooting via USB PD when it has access to one usb-c port#asahi · 2021-06-04 22:35
jannauinterrupts (AIC) are out of our control#asahi · 2021-06-04 22:29
jannauI'm not fully sure if I captured the exact moment before the disconnect#asahi · 2021-06-04 22:28
jannausven: https://gist.github.com/jannau/f24a41c69931e984e12221ed6950b255#asahi · 2021-06-04 22:13
jannausven: for which ADT nodes do you want mmiotraces?#asahi · 2021-06-04 22:01
jannauso slow :( "panic(cpu 0 caller 0xfffffe0014186ce4): "Halt/Restart Timed Out""#asahi · 2021-06-04 21:22
jannaueven dragging windows over the desktop is bearable but a little laggy from time to time#asahi · 2021-06-04 21:21
jannauI think so, fast enough interactive keyboard / mouse usage#asahi · 2021-06-04 21:19
jannauthe guest is noticable slower now that I deliver pmu irqs#asahi · 2021-06-04 21:18
jannaubug fixed but we probably have to account for cycles spent in the hypervisor#asahi · 2021-06-04 21:01
jannausecond usb-c port is useable in mac os but my perf monitor irq handling is broken #asahi · 2021-06-04 20:54
jannauthe hook for 0x23d280090 made it slow#asahi · 2021-06-04 20:49
jannaujust hiding one usb-c seems to work but my current PM irq changes seem to cause a FIQ storm in mac os#asahi · 2021-06-04 20:45
jannaumarcan: you mean pipcet[m] re dual booting#asahi · 2021-06-04 20:31
jannaumarcan: except that we currently hide both usb-c ports and need one for mmiotrace#asahi · 2021-06-04 07:14
jannaubgb: https://www.keystone-engine.org/#asahi · 2021-06-04 07:13
jannauI think I have to reset counters. clearing IACT probably does not work. macos seems to be fast enough to reset them so I get only a single irq no matter what I do#asahi · 2021-06-04 05:39
jannaurebooting by command is very convenient compared to pressing the power button#asahi · 2021-06-03 13:13
jannaumarcan: do they work now? i.e. is the guest's FIQ handler invoked#asahi · 2021-06-03 13:12
jannaubloom: yes. see https://github.com/AsahiLinux/docs/wiki/SW:Hypervisor#asahi · 2021-06-03 12:01
jannaumarcan: just not enaling PMC irqs and lying to the guest about it works as well#asahi · 2021-06-03 09:24
jannauI'll check. currently I'm shadowing bit 48 for each counter#asahi · 2021-06-03 09:23
jannauinterrupts are generated as long as bit 48 is set#asahi · 2021-06-03 09:22
jannauiirc the counters itself have be wrapped#asahi · 2021-06-03 09:14
jannauAIC IRQs are not handled by the hypervisor#asahi · 2021-06-03 09:13
jannauFIQ#asahi · 2021-06-03 09:12
jannauI'll prepare a PR#asahi · 2021-06-03 09:12
jannautrap the PMC msr and wrap the counters on IRQs#asahi · 2021-06-03 09:10
jannauthat said, I handle PMC interrupts now in a way to shutdown mac os properly#asahi · 2021-06-03 09:06
jannaumore like incomplete. there is a watchdog at 0x23d2b0000 / 0x23d2b8000 https://github.com/AsahiLinux/docs/wiki/HW:WDT#asahi · 2021-06-03 09:03
jannaucould be, the WDT in the adt is at 0x23d2b8000#asahi · 2021-06-03 09:02
jannauany idea what might be at mmio address 0x23d2b9000? it is not mentioned in the ADT as far as I can see. mac os writes up to offset 0x20 quite often. at offset 0x0 is counter and it writes increasing values to offset 0x4, 0x8, 0x10 and 0x20#asahi · 2021-06-02 05:29
jannauemilazy: u-boot provides UEFI and it is desireable so that distro's installers just work#asahi · 2021-06-02 05:26
jannaukernel updates only through 1TR or booting over uart/usb. The second has currently no security whatsoever#asahi · 2021-06-02 05:25
jannaumarcan: nothing wrong with booting Linux directly from m1n1. It is just inconvenient in production#asahi · 2021-05-31 20:10
jannauccache will be still helpful if you switch repeatedly between 4k and 16k#asahi · 2021-05-31 20:06
jannauuse ccache #asahi · 2021-05-31 19:38
jannauyes#asahi · 2021-05-31 19:38
jannaureadl_poll_timeout()#asahi · 2021-05-31 19:37
jannauI don't find that in the source#asahi · 2021-05-31 19:36
jannauwhen downloading a linux tarball over usb ethernet and writing it to usb disk