#
✕ nick:marcan
#asahi-dev · 2021-01-27 11:06
marcanah, so it just passes spintable info from the devicetree it gets?
#asahi-dev · 2021-01-27 11:05
marcanI haven't looked into any of this in u-boot yet
#asahi-dev · 2021-01-27 11:05
marcanalso depends on the spintable thing, whether u-boot can chain from that to the next stage or whether it reimplements it and thus needs its own CPU init code
#asahi-dev · 2021-01-27 11:04
marcanwell right now u-boot can't talk to the disk anyway, so not sure how much benefit that would be other than for fun :)
#asahi-dev · 2021-01-27 11:03
marcanbut to a first approximation we could leave all of m1n1 in reserved RAM, it's not like it's big anyway
#asahi-dev · 2021-01-27 11:03
marcanthough obviously that does not include the FS code
#asahi-dev · 2021-01-27 11:03
marcanhas to do with CPU init chicken bits and the resume vector being locked down
#asahi-dev · 2021-01-27 11:03
marcanand actually I think at least a portion of m1n1 needs to stay in RAM after OS boot anyway, to handle resume from sleep scenarios, though that part isn't clear yet
#asahi-dev · 2021-01-27 11:02
marcanm1n1 is relocatable and basically standalone, so I don't see why you couldn't just call into its disk I/O as long as you have some sort of sane stack going on
#asahi-dev · 2021-01-27 11:02
marcanonce I have the disk I/O code in here we can revisit, but basically if there is a way I can load your bootloader and give you hooks to get disk I/O - be it via UEFI or some bespoke mechanism - without it being too much useless overhead, that works perfectly fine for me
#asahi-dev · 2021-01-27 11:01
marcansure
#asahi-dev · 2021-01-27 11:00
marcanno problem then
#asahi-dev · 2021-01-27 11:00
marcanin there you can do whatever you want
#asahi-dev · 2021-01-27 10:59
marcanso if you want disklabels, they'll have to be inside a GPT partition
#asahi-dev · 2021-01-27 10:59
marcanapple requires GPT and APFS volumes to boot from
#asahi-dev · 2021-01-27 10:59
marcanobviously that's not going to fly on M1
#asahi-dev · 2021-01-27 10:58
marcanthe only reason I like FAT32 is that elm-chan fatfs is just so damn easy to use and works everywhere
#asahi-dev · 2021-01-27 10:58
marcanas long as the licensing is not a problem
#asahi-dev · 2021-01-27 10:57
marcanI'm not opposed to adding FS support as long as it's not NTFS levels of junk
#asahi-dev · 2021-01-27 10:57
marcanalso, if there is a simple enough read-only UFS2/etc implementation I can drop into m1n1 that's fine too
#asahi-dev · 2021-01-27 10:57
marcanfair
#asahi-dev · 2021-01-27 10:56
marcanjust UEFI?
#asahi-dev · 2021-01-27 10:56
marcanhow does openbsd get loaded in the u-boot path?
#asahi-dev · 2021-01-27 10:56
marcaneither by making m1n1 do enough of UEFI for you guys, if it's really a minimal part; or by chaining u-boot and teaching it about disk I/O
#asahi-dev · 2021-01-27 10:56
marcanand then for production, decide on how you want to hook up the support for loading kernels the normal way
#asahi-dev · 2021-01-27 10:56
marcanin that case it looks like for testing you could and probably should get away with loading openbsd kernels into m1n1, even via serial like I do for linux for development
#asahi-dev · 2021-01-27 10:55
marcanyeah, no offense but I'm not touching that thing
#asahi-dev · 2021-01-27 10:55
marcan:)
#asahi-dev · 2021-01-27 10:55
marcan(I'm going to write support for that in the next day or two)
#asahi-dev · 2021-01-27 10:54
marcanand what we will use a priori
#asahi-dev · 2021-01-27 10:54
marcanthat's the other option Linux has
#asahi-dev · 2021-01-27 10:54
marcancan you do the same spintable thing Linux supports?
#asahi-dev · 2021-01-27 10:53
marcanwhich is also not something I want to lose
#asahi-dev · 2021-01-27 10:53
marcanin that case it could be done, but then you lose virtualization support
#asahi-dev · 2021-01-27 10:53
marcanheh, didn't know hvc was an option
#asahi-dev · 2021-01-27 10:53
marcanaren't PSCI calls monitor calls?
#asahi-dev · 2021-01-27 10:52
marcanthere is no EL3
#asahi-dev · 2021-01-27 10:52
marcanso good luck with that
#asahi-dev · 2021-01-27 10:52
marcanPSCI needs trustzone, which does not exist on M1
#asahi-dev · 2021-01-27 10:52
marcan:)
#asahi-dev · 2021-01-27 10:52
marcanso we're already diverging there by necessity
#asahi-dev · 2021-01-27 10:52
marcanyeah, ARM also keeps pushing PSCI and there will never be PSCI on M1
#asahi-dev · 2021-01-27 10:51
marcanbut I certainly don't want to go down the path of deep hardware info stuff, ACPI, etc
#asahi-dev · 2021-01-27 10:51
marcanhow much of the UEFI API does it need? e.g. if all I need are protocol stubs for disk I/O (read/write sector) and console write (for text output to the framebuffer) then that could conceivably be added to m1n1
#asahi-dev · 2021-01-27 10:50
marcan(I've never used UEFI on ARM)
#asahi-dev · 2021-01-27 10:50
marcanah, heh
#asahi-dev · 2021-01-27 10:50
marcanhow does that fit in with UEFI - one or the other?
#asahi-dev · 2021-01-27 10:50
marcandoes openbsd support devicetrees?
#asahi-dev · 2021-01-27 10:49
marcan(I use it on a few machines)
#asahi-dev · 2021-01-27 10:49
marcanthere isn't a terribly good reason to use a fancy filesystem for the boot partition, e.g. this setup is one fairly common option on PCs (making /boot the ESP)

plus ›