#
✕ nick:marcan
#asahi-dev · 2022-01-06 13:25
marcannice :)
#asahi-dev · 2022-01-06 13:24
marcanif I'm lucky by the end of the month I might be able to get things to a near-testable point, and then I can also help out with any u-boot stuff that also needs doing, aiming for something releasable
#asahi-dev · 2022-01-06 13:24
marcanand then installer stuff
#asahi-dev · 2022-01-06 13:24
marcanI need to get SMC done and keep pushing the linux upstreaming wheel, update the asahi branch and then figure out what to do about the Mac Mini video problem in m1n1
#asahi-dev · 2022-01-06 13:24
marcanthat would be awesome!
#asahi-dev · 2022-01-06 13:23
marcanthat sounds like a good list
#asahi-dev · 2022-01-06 13:23
marcancool :)
#asahi-dev · 2022-01-06 13:13
marcandoes it serve any other purpose? on the Pro/Max it could enable the SD card reader and enable booting from that if u-boot can do sdhci-pci, but...
#asahi-dev · 2022-01-06 13:13
marcankettenis: huh, I somehow missed or forgot about that, heh
#asahi-dev · 2022-01-06 13:02
marcan(sorry for still being less than 100% around; I've ended up helping a friend with some personal stuff over this week and the next, but I'm still working on the wifi stuff/reviews and other kernel things on and off)
#asahi-dev · 2022-01-06 12:59
marcanjannau: oh, that's new :)
#asahi-dev · 2022-01-06 11:37
marcanlandscape15[m]: yes
#asahi-dev · 2022-01-06 11:37
marcanrobher: good catch, I'll send a fix for that
#asahi-dev · 2022-01-05 13:28
marcanI have an ALARM rootfs lying around now
#asahi-dev · 2022-01-05 13:14
marcanI guess I need to try iwd
#asahi-dev · 2022-01-05 11:46
marcanso it doesn't work, and this applies to all brcmfmac cards
#asahi-dev · 2022-01-05 11:46
marcanand it wants the host-side SAE
#asahi-dev · 2022-01-05 11:46
marcanwpa_supplicant does *not* support SAE offload
#asahi-dev · 2022-01-05 11:46
marcanit claims to support SAE offload (not non-offloaded SAE)
#asahi-dev · 2022-01-05 11:46
marcanI think the driver is fine?
#asahi-dev · 2022-01-05 11:46
marcaner, so about that WPA3 support thing
#asahi-dev · 2022-01-04 15:47
marcanwonder if that's apple's fault or broadcom's
#asahi-dev · 2022-01-04 15:47
marcanthat's an ugly one
#asahi-dev · 2022-01-04 15:47
marcankettenis: ouch
#asahi-dev · 2022-01-03 14:47
marcanlooks like they're basically feeding a bunch of metrics into PID controllers and using that to make scheduling and DVFS state decisions
#asahi-dev · 2022-01-03 14:43
marcanso now we know what CLPC is :)
#asahi-dev · 2022-01-03 14:43
marcanfound an interesting Apple patent: https://patents.google.com/patent/US10417054B2/en?oq=US10417054B2
#asahi-dev · 2022-01-02 14:50
marcanI need to do a cleanup pass on that three anyway once I merge the new stuff
#asahi-dev · 2022-01-02 14:50
marcankettenis: good enough :)
#asahi-dev · 2022-01-02 13:32
marcansure, please do :)
#asahi-dev · 2021-12-31 13:58
marcan42 bits
#asahi-dev · 2021-12-31 13:58
marcankettenis: t6000 does have a larger physical address space size, fwiw
#asahi-dev · 2021-12-31 10:08
marcanwouldn't surprise me if bypass mode just didn't work...
#asahi-dev · 2021-12-31 10:08
marcanI do know subpage protection is now mandatory
#asahi-dev · 2021-12-31 10:08
marcankettenis: I'm not sure we've tested bypass mode on t6000...
#asahi-dev · 2021-12-31 09:48
marcannice!
#asahi-dev · 2021-12-31 03:54
marcankettenis: https://github.com/AsahiLinux/m1n1/blob/main/proxyclient/m1n1/hv.py#L1057 turning that off should shut it up, dunno if it might break the single-step feature though
#asahi-dev · 2021-12-31 03:53
marcanthat's partially because the HV uses debug functionality, but we don't currently virtualize guest debug stuff properly anyway
#asahi-dev · 2021-12-31 03:53
marcankettenis: debug regs are basically the only standard regs we trap
#asahi-dev · 2021-12-27 10:54
marcanbut I find it more interesting how M1 Pro/Max configs for a given MBP (14" or 16") also have the same island; that *is* a different logic board, though presumably the WiFi section at least is 1:1 identical
#asahi-dev · 2021-12-27 10:53
marcanin that case I think the change is the display panel?
#asahi-dev · 2021-12-27 10:53
marcanit's not a 1:1 mapping to model IDs
#asahi-dev · 2021-12-27 10:53
marcanand yes the duplicate island is intentional
#asahi-dev · 2021-12-27 10:52
marcanI forgot to update that after I figured that one out
#asahi-dev · 2021-12-27 10:52
marcanrkjnsn[m]: ah yes, it's ACPI (x86)
#asahi-dev · 2021-12-26 16:22
marcanapple have their own NIH codebase though, in IOKit/C++; seems to share ~nothing with the broadcom code
#asahi-dev · 2021-12-26 16:22
marcanalso it's quite a bit less insane than bcmdhd ( the one I linked )
#asahi-dev · 2021-12-26 16:21
marcanhah, yeah
#asahi-dev · 2021-12-26 16:20
marcanfeel free to take whatever works!
#asahi-dev · 2021-12-26 16:20
marcanyup, I noticed that driver is ISC :)

plus ›