#
✕ nick:jannau
#asahi-dev · 2022-06-11 10:10
jannau2. prepare a m1n1 + u-boot bundle with smc and simple-framebuffer backlight disabled
#asahi-dev · 2022-06-11 10:09
jannau1. provide fixed u-boot and kernel packages
#asahi-dev · 2022-06-11 10:09
jannauI think fixup strategy for users with broken installs should be:
#asahi-dev · 2022-06-11 10:07
jannauhttps://github.com/jannau/linux/commit/6bf33007029983436b38c286c47420afdc022256
#asahi-dev · 2022-06-11 09:58
jannauone line patch for smc_rtkit.c
#asahi-dev · 2022-06-11 09:58
jannaubut yes, that's the issue
#asahi-dev · 2022-06-11 09:54
jannauor not, I just traced macos
#asahi-dev · 2022-06-11 09:52
jannausmc crashes with the macos 13 firmware because the linux driver acks the ioreport buffer request
#asahi-dev · 2022-06-11 09:26
jannaunow the smc crash
#asahi-dev · 2022-06-11 09:25
jannaumacos 13 u-boot fixes pushed to https://github.com/jannau/u-boot-1/tree/macos13-nvme - tested on t600x (macos 13) and t8103 (macos 12.3)
#asahi-dev · 2022-06-09 20:38
jannaukettenis: working state pushed to https://github.com/jannau/u-boot-1/tree/macos13-nvme with of based sart init, only tested on m1 max
#asahi-dev · 2022-06-09 20:27
jannauyes, chainloading m1n1-dcp-nb + dtb + linux should work with any previous m1n1
#asahi-dev · 2022-06-09 20:25
jannaupovik: not sure how well it rebases on asahi. m1n1 will be harder to rebase on main
#asahi-dev · 2022-06-09 20:23
jannauyou could try https://github.com/jannau/linux/tree/asahi-dcp-nb with https://github.com/jannau/m1n1/tree/dcp-nb
#asahi-dev · 2022-06-09 20:01
jannaumy current WIP state is improved and might work on all devices (only tested on m1 max so far though)
#asahi-dev · 2022-06-09 19:59
jannaupovik: for which device? my pushed branches probably work only for t8103 and I would have to check for the matching m1n1
#asahi-dev · 2022-06-09 19:26
jannaujust removing smc from the dt in the fixup m1n1 + u-boot bundle should be enough
#asahi-dev · 2022-06-09 19:25
jannaufixing existing installations with macos 13 beta is a little annoying due to crashing smc
#asahi-dev · 2022-06-09 19:20
jannauI think the issue was only not supplying buffers for crashlog (and ioreport). not necessary to start syslog
#asahi-dev · 2022-06-09 19:16
jannauyes
#asahi-dev · 2022-06-09 19:11
jannauand fixed
#asahi-dev · 2022-06-09 19:04
jannaukettenis: one issue in my nvme u-boot change found, the get buffer ack had size 0 since I used the size in bytes
#asahi-dev · 2022-06-09 05:23
jannaus/srt/sart/
#asahi-dev · 2022-06-09 05:23
jannauhttps://github.com/AsahiLinux/m1n1/pull/218 could be handy if you look at it
#asahi-dev · 2022-06-09 05:22
jannausrt just got me 1 additional rtkit message
#asahi-dev · 2022-06-09 05:22
jannaukettenis: I've pushed my nvme WIP state to https://github.com/jannau/u-boot-1/tree/macos13-nvme
#asahi-dev · 2022-06-08 17:14
jannaumarcan: linux macsmc driver crashes smc with macos 13-beta (m1 max)
#asahi-dev · 2022-06-07 21:53
jannaunot much progrees even with sart. I get a buffer request for IOREPORT (ep 4) but the mailbox is dead after that
#asahi-dev · 2022-06-07 19:57
jannauit looks like the issue is not providing buffers, it just stops after ack-ing the crashlog buffer request with dva=0
#asahi-dev · 2022-06-07 19:00
jannauu-boot nvme issue confirmed
#asahi-dev · 2022-06-07 16:56
jannauah, no surprise. I was just confused about model numbers
#asahi-dev · 2022-06-07 16:48
jannauis there a NDA on developer beta releases? the macos-13 beta contains a surprise
#asahi-dev · 2022-06-07 06:12
jannauI wonder what I have to do to support ProMotion. The mode DCP announces as 120 Hz (macOS reports it as ProMotion) seems to do only 60 swaps per second
#asahi-dev · 2022-06-07 06:05
jannaumaz: fixing an u-boot bug breaks on apple silicon https://lore.kernel.org/all/d3cdab5e8b174677@bloch.sibelius.xs4all.nl/
#asahi-dev · 2022-06-06 13:20
jannauplatform_get_irq might has to live in the apple specific part of the driver init
#asahi-dev · 2022-06-06 13:15
jannauamarioguy: you have to use the one specified in the ADT/FDT, not sure if we have transferred it to the device tree
#asahi-dev · 2022-06-05 15:36
jannauissue was unrelated to the buffer allocation
#asahi-dev · 2022-06-05 14:02
jannauthe annoying part is that dcp crashes after what I believe is the last re-allocation
#asahi-dev · 2022-06-05 14:00
jannauENODEV
#asahi-dev · 2022-06-05 13:45
jannaudcp is strange, on the internal display (mbp 14") a modeset (changing just the refresh rate) causes freeing and reallocating of buffers. That doesn't happen on mac mini or mac studio on mode changes
#asahi-dev · 2022-06-04 21:17
jannauno, :( - it failed just earlier so simpledrm kept working
#asahi-dev · 2022-06-04 21:15
jannauon the bright side: modesets (just changing the refresh rate) seem to work now on the macbook pro
#asahi-dev · 2022-06-04 21:13
jannauit was not that slow earlier. no on the macbook pro 14"
#asahi-dev · 2022-06-04 21:12
jannauand I only grabbed the phone and started recording after I saw it was slow
#asahi-dev · 2022-06-04 21:12
jannausigh, something is broken. I had now really glacial startup from disabling the MMU in m1n1. I recorded 2 min and 14 seconds
#asahi-dev · 2022-06-04 18:16
jannauand a little annoyed since modesets are crash dcp macbook but not on other devices
#asahi-dev · 2022-06-04 18:15
jannauis just surprised that the timing mode is saved in nvram
#asahi-dev · 2022-06-04 18:14
jannauthe display is initialized
#asahi-dev · 2022-06-04 18:13
jannauI don't think patching is necessary, iboot probably just configures it and it persists after that
#asahi-dev · 2022-06-04 18:09
jannauany idea how the macOS selected timing mode could persist in dcp over reboot and poweroff (macbook pro 14")?

plus ›