#asahi-dev · 2023-09-15 15:17
_jannau__aarch32 has no w registers. it uses rX mnemonics since all registers are 32-bit wide#asahi-dev · 2023-09-15 15:15
_jannau__most aarch64 instructions can operate on x and w register (aliases)#asahi-dev · 2023-09-15 14:53
_jannau__nicolas17: on one hand possible, ~10% faster but almost 10% max perf core frequency, 20% faster GPU after updating from 5 to 6 GPU cores. But I doubt apple would have talked about wider decode/execution units in the presentation if the cpu core were identical#asahi-dev · 2023-09-15 14:03
_jannau__should prepare the initramfs hooks to copy kernel modules into a tmpfs for the live system#asahi-dev · 2023-09-15 13:58
_jannau__you can also append an initrd#asahi-dev · 2023-09-15 13:57
_jannau__yes, in that case you probably want to append boot args as well#asahi-dev · 2023-09-15 13:56
_jannau__s/u-boot/next stage payload (usually u-boot)/#asahi-dev · 2023-09-15 13:54
_jannau__I think asahi alarm installs them in the module dir#asahi-dev · 2023-09-15 13:53
_jannau__they are embedded in ESP/m1n1/boot.bin, location in the system is distro specific, check the update-m1n1 source and/or config#asahi-dev · 2023-09-15 12:01
_jannau__probably with m1n1 installed in proxy only mode in a separate partition or m1n1 installed in a macOS partition for running macOS under the HV for tracing#asahi-dev · 2023-09-15 11:59
_jannau__for testing kernels most of us probably use something along https://github.com/AsahiLinux/docs/wiki/Tethered-Boot-Setup-(For-Developers)#asahi-dev · 2023-09-14 14:06
_jannau__looks more if there could be too tolerance on the mounting holes than out of spec tabs on the connector. at least sll the tabs are in their holes. I'll take photos of the worst offender#asahi-dev · 2023-09-14 10:17
_jannau__maz: boards have arrived. seems like jlbpcb has issues with mounting the micro usb ports straight. around half are visually off. just a quick visual inspection for now#asahi-dev · 2023-09-13 16:52
_jannau__see tools/reset_panic_counter.py#asahi-dev · 2023-09-13 16:52
_jannau__m1n1 can take care of that and it's taken care of automatically when using the HV#asahi-dev · 2023-09-13 16:36
_jannau__I find it useful to have a separate proxy only install on CI / test devices. that avoid the "Enabling the backdoor proxy mode" step#asahi-dev · 2023-09-13 16:30
_jannau__the hw reset needs some usb pd hardware#asahi-dev · 2023-09-13 16:29
_jannau__yes, kernel can be booted over usb (either barebone or with m1n1's HV which virtualizes mostly the serial port)#asahi-dev · 2023-09-13 16:27
_jannau__I should get a few populated cs-hw boards tomorrow#asahi-dev · 2023-09-13 16:26
_jannau__I suppose for CI the hard reset over USB PD is useful#asahi-dev · 2023-09-13 16:25
_jannau__we still have to implement that on Linux#asahi-dev · 2023-09-13 16:24
_jannau__yes, has to be enabled via USB PD, 1.2V#asahi-dev · 2023-09-13 16:24
_jannau__see https://github.com/AsahiLinux/vdmtool or https://git.kernel.org/pub/scm/linux/kernel/git/maz/cs-hw.git/#asahi-dev · 2023-09-13 16:23
_jannau__eric_engestrom: there is a serial console on the USB-C port next to the magsafe port#asahi-dev · 2023-09-08 10:07
_jannau__not sure if we already use it somewhere#asahi-dev · 2023-09-08 10:06
_jannau__we added dart dts properties to limit the aperture as well. basicly the same as vm-base, vm-size in the adt#asahi-dev · 2023-09-07 16:02
_jannau__have there been any surprise differences between Pro and MAx variants so far? I can't think of anything#asahi-dev · 2023-09-07 16:00
_jannau__are t6000 and t6001 identical?#asahi-dev · 2023-09-07 15:58
_jannau__same same#asahi-dev · 2023-09-07 15:58
_jannau__and not allowed to fix past mistakes#asahi-dev · 2023-09-07 15:57
_jannau__different teams not allowed to know of each other without adult oversight#asahi-dev · 2023-09-07 15:54
_jannau__is there a need to map specific things at specific iova? or is it needed due to dart mess? i.e. mapping subsets but only consistantly over multiple darts?#asahi-dev · 2023-09-07 13:28
_jannau__have you checked if i-boot creates a carveout? check carveout-memory-map in the adt#asahi-dev · 2023-09-04 08:14
_jannau__it's probably all user space. see also https://lore.kernel.org/lkml/7f2d88de-60c5-e2ff-9b22-acba35cfdfb6@redhat.com/#asahi-dev · 2023-09-04 08:11
_jannau__https://lore.kernel.org/dri-devel/b61d3eeb-6213-afac-2e70-7b9791c86d2e@redhat.com/ would be a possible solution but as far as I'm aware there have been no progress this year (except maybe still backlight device cleanup of existing drivers)#asahi-dev · 2023-09-04 08:09
_jannau__yes, the problem is that we are about to enable support for more than 1 display#asahi-dev · 2023-09-01 09:48
_jannau__eiln: the m1 pro/max macbook pros have a 1920x1080 sensor: https://support.apple.com/kb/SP854?locale=en_US#asahi-dev · 2023-08-30 15:44
_jannau__I expected something more advanced than a 7 year old iphone (at the time of the m1 release)#asahi-dev · 2023-08-30 15:41
_jannau__eiln: ah. are you using single buffer with page aligned luma and chroma?#asahi-dev · 2023-08-30 15:40
_jannau__a quick search suggest macos uses only 25 or 30 fps so might be an limitation of the sensor#asahi-dev · 2023-08-30 15:23
_jannau__does the single planar api uses YUYV/YUV2? any idea why you only get 33fps? is that a hw/fw/config issue or is it related to the single planar format? I would have expected that the camera/isp delivers 60fps#asahi-dev · 2023-08-30 15:19
_jannau__eiln: \o/ great. If you think a PR is fine then create a PR. we've certainly merged things for testing which still had issues#asahi-dev · 2023-08-30 09:41
_jannau__no worries, it would have been a convenient method to recover the fedora installs with broken grub configuration if it had worked#asahi-dev · 2023-08-30 09:38
_jannau__I don't know. it's the 3rd instruction according to gdb's disassembly but I don't really trust that. the two previous instructions are nops#asahi-dev · 2023-08-23 14:18
_jannau__did the one working on the OpenBSD not talk to kettenis?#asahi-dev · 2023-08-23 13:52
_jannau__err, no, that is not the driver#asahi-dev · 2023-08-23 13:51
_jannau__this one looks current https://android.googlesource.com/platform/hardware/broadcom/wlan/+/refs/heads/main#asahi-dev · 2023-08-23 13:49
_jannau__there might be vendor trees with less ancient versions#asahi-dev · 2023-08-23 13:47
_jannau__tobhe: yes, it's bcmdhd but I'm not sure if there a clear upstream source. there is however https://android.googlesource.com/kernel/common/+/bcmdhd-3.10/drivers/net/wireless/bcmdhd#asahi-dev · 2023-08-22 07:31
_jannau__mps: it should not. uinput has nothing to do with the touchscreen it is just a requiremnt of tiny-dfr