#
✕ nick:marcan
#asahi · 2022-12-04 07:33
marcanit says 12.6.1 should match what I detect as 12.5
#asahi · 2022-12-04 07:33
marcanm1n1 infers system firmware version from iBoot version and I used the table on the iPhone wiki
#asahi · 2022-12-04 07:33
marcanyou can try running the asahi-installer and pasting the sys info blurb
#asahi · 2022-12-04 07:30
marcanAre you *sure*?
#asahi · 2022-12-04 07:29
marcan:D
#asahi · 2022-12-04 07:28
marcanwonder if someone leaked the iBoot keys for 13.0 on t8103...
#asahi · 2022-12-04 07:28
marcanah cool, 13.0, maybe they added that message or something
#asahi · 2022-12-04 07:25
marcanit is, if you just updated, asahi-scripts was the last thing that went in a bit earlier
#asahi · 2022-12-04 07:23
marcanyeah
#asahi · 2022-12-04 07:14
marcanspeaking of, we really should start providing some kind of `asahi-diagnose` script that just collects All The Info and requiring it for bug reports
#asahi · 2022-12-04 07:14
marcanadded a bunch of goodies :)
#asahi · 2022-12-04 07:13
marcannicolas17: update asahi-dev, reboot, and try `grep -a . /proc/device-tree/chosen/*-version`
#asahi · 2022-12-04 07:01
marcannicolas17: not yet, kind of got carried away also implementing the DCP/GPU firmware versioning while I was here, almost done
#asahi · 2022-12-04 06:04
marcanthanks :)
#asahi · 2022-12-04 06:03
marcanhttps://wiki.archlinux.org/title/sway#HiDPI
#asahi · 2022-12-04 06:02
marcanenable HiDPI support
#asahi · 2022-12-04 06:01
marcanblackMysticCat: you cannot change the resolution on internal displays (they do not support it).
#asahi · 2022-12-04 06:01
marcannicolas17: it is :)
#asahi · 2022-12-04 05:47
marcanmaybe the UI should know about upower settings and display something like "UPower is configured to perform a critical shutdown at 2%"
#asahi · 2022-12-04 05:46
marcanit makes sense for there to be a forced last-measure shutdown
#asahi · 2022-12-04 05:46
marcanyour computer is going to shut down when the battery dies whether you want it to or not
#asahi · 2022-12-04 05:46
marcanwell, it was at 2%
#asahi · 2022-12-04 05:42
marcanit's important stuff to have access to
#asahi · 2022-12-04 05:42
marcangive me a sec, I'll add this quickly and push out a release ;)
#asahi · 2022-12-04 05:41
marcanyou can from m1n1, but let me just add something for this into kboot real quick because we're going to want it handy
#asahi · 2022-12-04 05:37
marcan(I should make m1n1 forward this stuff to the DT...)
#asahi · 2022-12-04 05:36
marcannicolas17: what's your SFR? (i.e. newest version of macOS ever installed)
#asahi · 2022-12-04 05:36
marcanI don't recall that syslog at all from my SMC experiments, maybe it's new?
#asahi · 2022-12-04 05:36
marcanthough I still want to find an explicit event I can use to trigger a panic kernel flush
#asahi · 2022-12-04 05:36
marcanlooks like the current behavior is "good enough" then, if it shuts down at 2%
#asahi · 2022-12-04 05:34
marcanand yet you had 10 minutes after SMC reported, apparently, "battery too low to boot". So maybe that state is indeed higher than 0% for macOS
#asahi · 2022-12-04 05:34
marcanit's interesting though, because I think we use the "UI capacity" to report battery %, and I'd expect Apple to let that drop to 0% before initiating shutdown (even though there's still reserve power available)
#asahi · 2022-12-04 05:33
marcanthat makes sense
#asahi · 2022-12-04 05:32
marcan(and then later the panic one I found from the kernel)
#asahi · 2022-12-04 05:32
marcanbecause that condition is exactly what I'd want to use to trigger a clean shutdown from userspace
#asahi · 2022-12-04 05:32
marcanbut lasting 10 minutes after that is more than I expected, I wonder if there's some event/prop I missed there
#asahi · 2022-12-04 05:31
marcannicolas17: the okBootiBoot=1/okBootAP_OS=0 thing is interesting, it suggests SMC thinks the battery is now too dead to actually boot the OS, which is an interesting state to be in
#asahi · 2022-12-04 05:25
marcanthat sounds like something userspace, yeah
#asahi · 2022-12-04 05:19
marcanit would be nice if there's some flag we can enable to actually get an event delivered on battery critical
#asahi · 2022-12-04 05:18
marcanwould be nice to know how a lot of this works (and looking at firmware is fairly safe from a cleanroom perspective, since you're just looking at the "other side" of what you have to implement)
#asahi · 2022-12-04 05:18
marcanI did look at some iOS SMC firmwares before, but they're missing a lot of stuff that's in the macs
#asahi · 2022-12-04 05:18
marcanI should throw SMC into ghidra, now that the t8103 iboot encryption keys leaked for some firmwares
#asahi · 2022-12-04 05:17
marcaninstead of waiting for it to kill power itself
#asahi · 2022-12-04 05:17
marcanSMC gives us three warnings, we do a software initiated hard shutdown on the third (which flushes NVMe)
#asahi · 2022-12-04 05:17
marcanwe already have panic shutdown when you hold the power button, FWIW
#asahi · 2022-12-04 05:16
marcan(on desktops I'm not sure there's much we can do, but there's a kernel arg for those who want to disable the defers at the cost of significant flush performance)
#asahi · 2022-12-04 05:15
marcanbut on laptops it should be possible to close it entirely
#asahi · 2022-12-04 05:15
marcanI find that unacceptable, right now our window is much smaller than macOS though we still have it on desktops
#asahi · 2022-12-04 05:14
marcane.g. if you rsync data over then delete the source, pull the plug, data is gone from both sides
#asahi · 2022-12-04 05:14
marcanFWIW macOS has this problem, at least on desktops if you pull the plug

plus ›