#
✕ nick:jannau
#asahi · 2022-03-29 18:51
jannauother 3 were working but the shared irq got disabled
#asahi · 2022-03-29 18:50
jannaupovik: 3 of 3 non-working fixed
#asahi · 2022-03-29 17:45
jannaucpu frequency scaling seems to be busted on the ultra, so that means the perf cores are just running just at 2GHz
#asahi · 2022-03-29 17:42
jannauI wouldn't assume that 1284 is the final m1 ultra single core score
#asahi · 2022-03-29 17:35
jannauthe ultra got 15k
#asahi · 2022-03-29 16:42
jannauthis discussion is drifting strongly into off-topic
#asahi · 2022-03-29 16:31
jannauChainfire: it's geekbench issue. kernel compiles are on the m1 ultra twice as fast as on the m1 max
#asahi · 2022-03-29 16:28
jannausingle core results will be lower on asahi since we con't yet run a single core at 3.2GHZ
#asahi · 2022-03-29 16:27
jannautoo heavy
#asahi · 2022-03-29 16:26
jannauload avg barely reached 9
#asahi · 2022-03-29 16:25
jannauthe multicore test doesn't scale well, here are results from a m1 ultra: https://browser.geekbench.com/v5/cpu/13932507
#asahi · 2022-03-29 16:24
jannauj`ey posted that already, I was looking under downloads
#asahi · 2022-03-29 16:10
jannaubluetail[m]: it looks like there are just x86 binaries for linux
#asahi · 2022-03-29 16:04
jannaupovik, sven: "i2c: pasemi: Wait for write xfers to finish" fixes tps6598x probe errors (-EIO) for 3 of the 6 usb-c ports on the mac studio when using modules and/or booting through u-boot
#asahi · 2022-03-28 21:19
jannauj`ey: 155.56 seconds with a single m1 max
#asahi · 2022-03-28 20:47
jannauj`ey: m1 max was iirc very close to factor 2, I don't have numbers for m1
#asahi · 2022-03-28 17:15
jannaumarcan: yes, it is also not worth investing any effort in adapting m1n1's fb code to 8-bit, time would be better spent on the dcp driver
#asahi · 2022-03-28 17:06
jannauat the moment yes
#asahi · 2022-03-28 15:10
jannauGlanzmann: rtkit timeout for 4 and 8 TB SSDs and the mem layout fix
#asahi · 2022-03-28 13:45
jannauif so I suggest you try to work on that. we son't really have well defined and ready tasks beyond add support for $HARDWARE_PART
#asahi · 2022-03-28 13:44
jannausjbtrn[m]: welcome and nice that you want to contribute. Do you have already a apple silicon computer which runs asahilinux? if so is there anything you're missing?
#asahi · 2022-03-28 13:20
jannauit's the format iboot uses and the framebuffer has when m1n1 takes over. we had so far no way to switch it
#asahi · 2022-03-28 13:10
jannauj`ey: weston is ok at 4k simpledrm though
#asahi · 2022-03-28 13:01
jannauGlanzmann: the studio display will not work. this is just hdmi and not thunderbolt. in addition its resolution is 5120x2880 which would not fit
#asahi · 2022-03-28 12:53
jannauI swap now things around which will work up to 5120x2252
#asahi · 2022-03-28 12:53
jannauthe problem was that new framebuffer was partially in a new dart level 2 page table which vanishes at m1n1 exit
#asahi · 2022-03-28 12:51
jannauj`ey: slow, just console, haven't tried X11/wayland yet
#asahi · 2022-03-28 12:46
jannauGlanzmann: display_fb_alloc updated with working 4k framebuffer
#asahi · 2022-03-27 19:53
jannauGlanzmann: 4k is still roughly truncated after 11 / 15 of the height. which is good match for the preallocated fb size of 0x1500000
#asahi · 2022-03-27 19:37
jannau4k does however not work
#asahi · 2022-03-27 19:37
jannauGlanzmann: I can't reproduce anymore. problem was entirely on my side
#asahi · 2022-03-27 18:36
jannaumine is based on df53f16f763ef83c 19th march
#asahi · 2022-03-27 18:35
jannauI don't think that updating will help
#asahi · 2022-03-27 18:30
jannauGlanzmann: I can confirm the black linux screen. that seems to be unrelated to my changes. I see that with the stage 1 m1n1 from the installer (not recent) and linux.py
#asahi · 2022-03-27 09:40
jannauok
#asahi · 2022-03-27 09:40
jannauI haven't checked yet, it seemed so obvious that it should be the frame buffer that I ever investigated
#asahi · 2022-03-27 09:39
jannauI don't think I've ever compared it to the framebuffer base in bootargs
#asahi · 2022-03-27 09:37
jannaumarcan: /vram is mapped in dcp/disp0 dart though
#asahi · 2022-03-27 09:30
jannauI haven't looked yet, the annotated "/vram" is too small
#asahi · 2022-03-27 09:28
jannauit's certainly too small for 1920x1200
#asahi · 2022-03-27 09:27
jannauno problem
#asahi · 2022-03-27 09:26
jannauand on the mac studio it is just large enough for 1920x1080
#asahi · 2022-03-27 09:25
jannaumarcan: argh, I started working on the same functionality, I have a proptoype for allocating a larger framebuffer. it will be too small for 4k on the mac mini
#asahi · 2022-03-26 20:21
jannauFayn: https://www.openssh.com/history.html
#asahi · 2022-03-26 19:19
jannausee https://youtu.be/1zuULubtdQk?t=5397 for example
#asahi · 2022-03-26 19:18
jannaumatde[m]: the design in the stream was stm32, fusb302, level shifters and an usb hub
#asahi · 2022-03-26 19:16
jannauj`ey: I thnk there were only "schemantics" in inkscape
#asahi · 2022-03-26 19:14
jannaumatde[m]: I think there's not much besides following youtube stream: https://www.youtube.com/watch?v=1zuULubtdQk
#asahi · 2022-03-26 19:11
jannaunot impossible though, proxy would have worked with just the m1 max usb ports
#asahi · 2022-03-26 19:09
jannauj`ey: for the most parts, HW bringup still requires the HW uart. Mac studio bringup would have been a lot harder without it

plus ›